Include the baselines, comparisons and study documents, not just the fork

The packaging commit shipped the code and a summary of the numbers. It left out everything the numbers are measured *against*, which is most of what makes them checkable. Added: baselines/ llama.cpp runs (thread sweeps, NUMA sweeps, the interleaved A/B passes behind the headline decode figure) and the quality references: each build against its own bf16 over 12 greedy prompts. benchmarks/.../results/ 102 artifacts backing the published figures, each with a provenance block naming the model, quantization, thread set and build SHAs. docs/engines/ the engine analyses -- llama.cpp, ExecuTorch, MNN, ncnn, vLLM, vLLM-Omni -- and the cross-engine and device-simulation comparisons. docs/analysis/ the full study documents the summaries distil. docs/edge/comparisons.md where this fork sits against llama.cpp and why: decode is llama.cpp's, prefill is ours by ~2x, and since neither engine exceeds 39% of the measured 222 GB/s ceiling the decode gap is the execution model rather than the kernels. docs/edge/environment.md pins and machine, and the correction below. On "the modified vllm": there isn't one, and the doc says so plainly because it is the obvious assumption. The vLLM checkout is clean upstream (0 commits ahead, 0 uncommitted) and the installed wheel is stock 0.28.0+cpu. This fork changes vLLM's CPU *behaviour* through vLLM's own extension points at import time -- three CPU patches in vllm_omni/patch.py, the quantization-config registry, the IR op registry, platform hooks -- specifically so the changes survive a wheel upgrade rather than being lost on it. One of those patches is why "KV sizing is not a memory lever" looked true for months: CpuPlatform overwrote kv_cache_memory_bytes unconditionally, so the setting had no effect. There is no patch series to apply; pin the version and let patch.py do the rest. .gitignore gains explicit exceptions: upstream's blanket `*.json` and `results/` rules were silently eating the measurement artifacts, which in this fork are the substance rather than build output. Binary captures stay ignored. No vLLM or llama.cpp source is redistributed -- only output produced with them. NOTICE records both. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L1ax5wYLA6XbqEL1i8L9gZ
2026-09-14 14:01:12 +08:00
1 parent b36a3431bd
commit 9264ff81ab
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:59:20Z",
+    "avg_ns": 580079289,
+    "stddev_ns": 22445145,
+    "avg_ts": 883.695051,
+    "stddev_ts": 34.173372,
+    "samples_ns": [ 550140411, 579685176, 611186497, 588372185, 571012179 ],
+    "samples_ts": [ 930.671, 883.238, 837.715, 870.197, 896.653 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:59:24Z",
+    "avg_ns": 2118395683,
+    "stddev_ns": 31881177,
+    "avg_ts": 60.434035,
+    "stddev_ts": 0.909662,
+    "samples_ns": [ 2110731151, 2152277901, 2078510078, 2149464130, 2100995155 ],
+    "samples_ts": [ 60.6425, 59.4719, 61.5826, 59.5497, 60.9235 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:50:11Z",
+    "avg_ns": 453009837,
+    "stddev_ns": 4617262,
+    "avg_ts": 1130.311918,
+    "stddev_ts": 11.483962,
+    "samples_ns": [ 459248279, 451575469, 456242077, 449875010, 448108352 ],
+    "samples_ts": [ 1114.87, 1133.81, 1122.21, 1138.09, 1142.58 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:50:14Z",
+    "avg_ns": 1475847924,
+    "stddev_ns": 9833345,
+    "avg_ts": 86.732884,
+    "stddev_ts": 0.578067,
+    "samples_ns": [ 1489621433, 1475111972, 1476264136, 1476396077, 1461846002 ],
+    "samples_ts": [ 85.9279, 86.7731, 86.7054, 86.6976, 87.5605 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:44:52Z",
+    "avg_ns": 487743656,
+    "stddev_ns": 25108865,
+    "avg_ts": 1051.969118,
+    "stddev_ts": 54.409083,
+    "samples_ns": [ 484545979, 492513415, 522973237, 486232711, 452452942 ],
+    "samples_ts": [ 1056.66, 1039.57, 979.018, 1052.99, 1131.61 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T18:44:55Z",
+    "avg_ns": 1589326576,
+    "stddev_ns": 47473222,
+    "avg_ts": 80.594928,
+    "stddev_ts": 2.413871,
+    "samples_ns": [ 1529730792, 1557930160, 1636596682, 1585818829, 1636556420 ],
+    "samples_ts": [ 83.6749, 82.1603, 78.2111, 80.7154, 78.213 ]
+  }
+]
@@ -0,0 +1,106 @@
+"""Greedy-decode parity for llama.cpp against the same fp32 reference vLLM is
+scored on.
+
+The point of the comparison is that "4-bit" is not one thing: llama.cpp's
+Q4_K_M and our `cpu_int4` spend a similar number of bits per weight but pick
+different grids, so the speed comparison is only meaningful next to a quality
+comparison run on identical prompts.  llama-server takes a prompt as raw token
+ids, so both engines see exactly the token sequence the reference did.
+"""
+import argparse, json, subprocess, time, urllib.request
+
+import numpy as np
+
+
+def post(url, payload, timeout=600):
+    req = urllib.request.Request(
+        url, data=json.dumps(payload).encode(),
+        headers={"Content-Type": "application/json"})
+    with urllib.request.urlopen(req, timeout=timeout) as r:
+        return json.loads(r.read())
+
+
+def main():
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--gguf", required=True)
+    ap.add_argument("--ref", default="ref.json")
+    ap.add_argument("--ref-logits", default="ref_logits.npy")
+    ap.add_argument("--bin", default="/data/zhoutaichang/embedding_infer/llama.cpp/build/bin/llama-server")
+    ap.add_argument("--cores", default="56-87")
+    ap.add_argument("--threads", type=int, default=32)
+    ap.add_argument("--port", type=int, default=18099)
+    ap.add_argument("--ctx", type=int, default=4096)
+    ap.add_argument("--out", default="parity_llamacpp.json")
+    a = ap.parse_args()
+
+    ref = json.load(open(a.ref))
+    ref_logits = np.load(a.ref_logits)
+    n_new = len(ref["new_ids"])
+
+    proc = subprocess.Popen(
+        ["taskset", "-c", a.cores, a.bin, "-m", a.gguf, "--port", str(a.port),
+         "-t", str(a.threads), "-c", str(a.ctx), "--no-warmup", "--host", "127.0.0.1"],
+        stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
+    url = f"http://127.0.0.1:{a.port}"
+    try:
+        for _ in range(180):
+            try:
+                urllib.request.urlopen(url + "/health", timeout=2)
+                break
+            except Exception:
+                time.sleep(1)
+        else:
+            raise RuntimeError("llama-server did not become healthy")
+
+        res = post(url + "/completion", {
+            "prompt": ref["prompt_ids"], "n_predict": n_new,
+            "temperature": 0.0, "top_k": 1, "n_probs": 20,
+            "cache_prompt": False, "post_sampling_probs": False,
+        })
+    finally:
+        proc.terminate()
+        try:
+            proc.wait(timeout=30)
+        except subprocess.TimeoutExpired:
+            proc.kill()
+
+    probs = res.get("completion_probabilities") or []
+    ids = [p["id"] for p in probs] if probs and "id" in probs[0] else []
+    n = min(n_new, len(ids)) if ids else 0
+    match = sum(int(x == y) for x, y in zip(ref["new_ids"][:n], ids[:n]))
+
+    rep = {
+        "engine": "llama.cpp", "gguf": a.gguf.split("/")[-1],
+        "ref_text": ref["text"], "llamacpp_text": res.get("content", ""),
+        "greedy_token_match": f"{match}/{n}" if n else "n/a",
+        "exact_match": bool(n) and ref["new_ids"][:n] == ids[:n],
+        "ref_top1": ref["top1"],
+        "llamacpp_first_token": ids[0] if ids else None,
+        "first_token_agree": bool(ids) and ids[0] == ref["top1"],
+    }
+    if probs:
+        import math
+        top = (probs[0].get("top_logprobs") or probs[0].get("top_probs")
+               or probs[0].get("probs") or [])
+        first = {}
+        for e in top:
+            if "id" not in e:
+                continue
+            first[int(e["id"])] = (float(e["prob"]) if "prob" in e
+                                   else math.exp(float(e["logprob"])))
+        ref_top = np.argsort(-ref_logits)[:20]
+        ref_p = np.exp(ref_logits - ref_logits.max())
+        ref_p /= ref_p.sum()
+        pairs = [(int(t), float(ref_p[t]), first[int(t)]) for t in ref_top if int(t) in first]
+        rep["top20_prob_overlap"] = f"{len(pairs)}/20"
+        if pairs:
+            d = [abs(r - v) for _, r, v in pairs]
+            rep["top20_prob_max_abs_diff"] = max(d)
+            rep["top20_prob_mean_abs_diff"] = sum(d) / len(d)
+    print(json.dumps(rep, indent=2, ensure_ascii=False))
+    with open(a.out, "w") as f:
+        json.dump(rep, f, indent=2, ensure_ascii=False)
+
+
+if __name__ == "__main__":
+    main()
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:04:49Z",
+    "avg_ns": 489021600,
+    "stddev_ns": 23921490,
+    "avg_ts": 1048.939735,
+    "stddev_ts": 49.910229,
+    "samples_ns": [ 527122937, 485801934, 462083348, 490951261, 479148522 ],
+    "samples_ts": [ 971.31, 1053.93, 1108.03, 1042.87, 1068.56 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:04:52Z",
+    "avg_ns": 1570942064,
+    "stddev_ns": 20220055,
+    "avg_ts": 81.490699,
+    "stddev_ts": 1.061319,
+    "samples_ns": [ 1586700558, 1582528379, 1537750061, 1565679809, 1582051516 ],
+    "samples_ts": [ 80.6705, 80.8832, 83.2385, 81.7536, 80.9076 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:14:20Z",
+    "avg_ns": 464927159,
+    "stddev_ns": 22146638,
+    "avg_ts": 1103.200971,
+    "stddev_ts": 51.267874,
+    "samples_ns": [ 453191976, 479254490, 448568607, 496875740, 446744984 ],
+    "samples_ts": [ 1129.76, 1068.33, 1141.41, 1030.44, 1146.07 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:14:23Z",
+    "avg_ns": 1422234679,
+    "stddev_ns": 12240631,
+    "avg_ts": 90.004494,
+    "stddev_ts": 0.767183,
+    "samples_ns": [ 1422285965, 1416723437, 1414194914, 1414611159, 1443357923 ],
+    "samples_ts": [ 89.996, 90.3493, 90.5109, 90.4842, 88.6821 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:20:46Z",
+    "avg_ns": 456731353,
+    "stddev_ns": 6199988,
+    "avg_ts": 1121.174997,
+    "stddev_ts": 15.277320,
+    "samples_ns": [ 459406023, 462620142, 461507407, 450329050, 449794146 ],
+    "samples_ts": [ 1114.48, 1106.74, 1109.41, 1136.95, 1138.3 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 24,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T19:20:49Z",
+    "avg_ns": 1506056809,
+    "stddev_ns": 24193761,
+    "avg_ts": 85.007660,
+    "stddev_ts": 1.362252,
+    "samples_ns": [ 1478402259, 1532453276, 1530685126, 1494657866, 1494085519 ],
+    "samples_ts": [ 86.58, 83.5262, 83.6227, 85.6383, 85.6711 ]
+  }
+]
@@ -0,0 +1,272 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:32:06Z",
+    "avg_ns": 1498583726,
+    "stddev_ns": 7171856,
+    "avg_ts": 341.662152,
+    "stddev_ts": 1.628172,
+    "samples_ns": [ 1493166066, 1510351399, 1500283975, 1493415633, 1495701559 ],
+    "samples_ts": [ 342.896, 338.994, 341.269, 342.838, 342.314 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:32:18Z",
+    "avg_ns": 5062657211,
+    "stddev_ns": 186447127,
+    "avg_ts": 25.309483,
+    "stddev_ts": 0.893384,
+    "samples_ns": [ 4950579569, 4934896396, 5035243526, 5388250087, 5004316480 ],
+    "samples_ts": [ 25.8556, 25.9377, 25.4208, 23.7554, 25.5779 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:32:43Z",
+    "avg_ns": 560100709,
+    "stddev_ns": 15689530,
+    "avg_ts": 914.690155,
+    "stddev_ts": 25.391149,
+    "samples_ns": [ 566521408, 541347341, 583132866, 553334911, 556167020 ],
+    "samples_ts": [ 903.761, 945.788, 878.016, 925.299, 920.587 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:32:47Z",
+    "avg_ns": 1933840318,
+    "stddev_ns": 51036422,
+    "avg_ts": 66.225568,
+    "stddev_ts": 1.706730,
+    "samples_ns": [ 1896818220, 1923412956, 1937369806, 1892702916, 2018897692 ],
+    "samples_ts": [ 67.4814, 66.5484, 66.069, 67.6282, 63.4009 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:32:57Z",
+    "avg_ns": 635046952,
+    "stddev_ns": 3342779,
+    "avg_ts": 806.257449,
+    "stddev_ts": 4.239145,
+    "samples_ns": [ 634661823, 637237702, 639487443, 631413242, 632434552 ],
+    "samples_ts": [ 806.729, 803.468, 800.641, 810.879, 809.57 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:33:01Z",
+    "avg_ns": 1790378809,
+    "stddev_ns": 14069215,
+    "avg_ts": 71.496788,
+    "stddev_ts": 0.562779,
+    "samples_ns": [ 1803158642, 1794510633, 1803271959, 1776172087, 1774780727 ],
+    "samples_ts": [ 70.9865, 71.3286, 70.9821, 72.0651, 72.1216 ]
+  }
+]
@@ -0,0 +1,272 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:02Z",
+    "avg_ns": 1017103540,
+    "stddev_ns": 116210707,
+    "avg_ts": 509.454206,
+    "stddev_ts": 66.492143,
+    "samples_ns": [ 1045283058, 1099078888, 1103686903, 1017870763, 819598088 ],
+    "samples_ts": [ 489.819, 465.845, 463.9, 503.011, 624.696 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:08Z",
+    "avg_ns": 4481043455,
+    "stddev_ns": 53485519,
+    "avg_ts": 28.568042,
+    "stddev_ts": 0.342133,
+    "samples_ns": [ 4526208574, 4537101214, 4480636429, 4454908368, 4406362692 ],
+    "samples_ts": [ 28.2797, 28.2118, 28.5674, 28.7324, 29.0489 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:31Z",
+    "avg_ns": 343075990,
+    "stddev_ns": 8202776,
+    "avg_ts": 1493.062250,
+    "stddev_ts": 35.634573,
+    "samples_ns": [ 333407890, 348737139, 337165270, 342580383, 353489272 ],
+    "samples_ts": [ 1535.66, 1468.15, 1518.54, 1494.54, 1448.42 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:33Z",
+    "avg_ns": 1673245422,
+    "stddev_ns": 44193699,
+    "avg_ts": 76.540484,
+    "stddev_ts": 2.008808,
+    "samples_ns": [ 1727982320, 1650954337, 1650417455, 1711748955, 1625124045 ],
+    "samples_ts": [ 74.0748, 77.5309, 77.5561, 74.7773, 78.7632 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:42Z",
+    "avg_ns": 433558704,
+    "stddev_ns": 61442386,
+    "avg_ts": 1198.988860,
+    "stddev_ts": 159.663627,
+    "samples_ns": [ 384355143, 403569845, 520552799, 475384982, 383930753 ],
+    "samples_ts": [ 1332.1, 1268.68, 983.57, 1077.02, 1333.57 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:35:44Z",
+    "avg_ns": 1574917625,
+    "stddev_ns": 239068571,
+    "avg_ts": 82.814413,
+    "stddev_ts": 12.730560,
+    "samples_ns": [ 1383407468, 1295986814, 1572559050, 1776956669, 1845678125 ],
+    "samples_ts": [ 92.5252, 98.7664, 81.396, 72.0333, 69.3512 ]
+  }
+]
@@ -0,0 +1,272 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:19:27Z",
+    "avg_ns": 7115257366,
+    "stddev_ns": 607044890,
+    "avg_ts": 72.394533,
+    "stddev_ts": 6.405320,
+    "samples_ns": [ 7765939766, 6729867579, 7316853262, 7495192180, 6268434044 ],
+    "samples_ts": [ 65.9289, 76.0788, 69.9754, 68.3105, 81.6791 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:20:09Z",
+    "avg_ns": 30700265663,
+    "stddev_ns": 4064662246,
+    "avg_ts": 4.196201,
+    "stddev_ts": 0.381497,
+    "samples_ns": [ 32683898432, 32626033248, 27736648629, 27745563194, 32709184813 ],
+    "samples_ts": [ 3.9163, 3.92325, 4.61483, 4.61335, 3.91327 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:22:44Z",
+    "avg_ns": 4412049470,
+    "stddev_ns": 642897239,
+    "avg_ts": 118.057116,
+    "stddev_ts": 17.378814,
+    "samples_ns": [ 4619378132, 3608885475, 4557507327, 3988149647, 5286326771 ],
+    "samples_ts": [ 110.837, 141.872, 112.342, 128.38, 96.8536 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:23:12Z",
+    "avg_ns": 13160648883,
+    "stddev_ns": 4037164376,
+    "avg_ts": 9.845522,
+    "stddev_ts": 1.260348,
+    "samples_ns": [ 12727171856, 13397387109, 13869421316, 15023258698, 10786005440 ],
+    "samples_ts": [ 10.0572, 9.5541, 9.22894, 8.52012, 11.8672 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:24:19Z",
+    "avg_ns": 4955359522,
+    "stddev_ns": 278495888,
+    "avg_ts": 103.589056,
+    "stddev_ts": 5.933882,
+    "samples_ns": [ 4969396818, 4806604145, 5198757330, 4567278377, 5234760941 ],
+    "samples_ts": [ 103.031, 106.52, 98.4851, 112.102, 97.8077 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:24:49Z",
+    "avg_ns": 12648866627,
+    "stddev_ns": 4253557830,
+    "avg_ts": 10.355908,
+    "stddev_ts": 1.831931,
+    "samples_ns": [ 12076753955, 12558032815, 15171316078, 9645037747, 13793192540 ],
+    "samples_ts": [ 10.5989, 10.1927, 8.43697, 13.2711, 9.27994 ]
+  }
+]
@@ -0,0 +1,272 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:28:17Z",
+    "avg_ns": 2946090115,
+    "stddev_ns": 19481720,
+    "avg_ts": 173.795716,
+    "stddev_ts": 1.144798,
+    "samples_ns": [ 2976783425, 2943842928, 2936842142, 2948762947, 2924219135 ],
+    "samples_ts": [ 171.998, 173.922, 174.337, 173.632, 175.089 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B.gguf",
+    "model_type": "spark2_5 1.7B BF16",
+    "model_size": 3415547904,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:28:34Z",
+    "avg_ns": 7926507628,
+    "stddev_ns": 166210109,
+    "avg_ts": 16.153997,
+    "stddev_ts": 0.336782,
+    "samples_ns": [ 7746359669, 7836100448, 8141583693, 8059318007, 7849176324 ],
+    "samples_ts": [ 16.5239, 16.3347, 15.7218, 15.8822, 16.3074 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:29:15Z",
+    "avg_ns": 1044627273,
+    "stddev_ns": 8668625,
+    "avg_ts": 490.153690,
+    "stddev_ts": 4.025039,
+    "samples_ns": [ 1040367711, 1059967868, 1042470115, 1039018745, 1041311927 ],
+    "samples_ts": [ 492.134, 483.034, 491.141, 492.773, 491.687 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q8_0.gguf",
+    "model_type": "spark2_5 1.7B Q8_0",
+    "model_size": 1814728704,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:29:21Z",
+    "avg_ns": 2844703994,
+    "stddev_ns": 50284844,
+    "avg_ts": 45.007005,
+    "stddev_ts": 0.785784,
+    "samples_ns": [ 2841360663, 2852415251, 2800656934, 2925026017, 2804061106 ],
+    "samples_ts": [ 45.0488, 44.8743, 45.7036, 43.7603, 45.6481 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:29:36Z",
+    "avg_ns": 1179360692,
+    "stddev_ns": 9282489,
+    "avg_ts": 434.155009,
+    "stddev_ts": 3.414034,
+    "samples_ns": [ 1190805907, 1173513176, 1168400903, 1186783926, 1177299549 ],
+    "samples_ts": [ 429.961, 436.297, 438.206, 431.418, 434.894 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T06:29:43Z",
+    "avg_ns": 3018952734,
+    "stddev_ns": 27599225,
+    "avg_ts": 42.401651,
+    "stddev_ts": 0.388562,
+    "samples_ns": [ 3042267389, 2995063052, 3045385324, 2984815648, 3027232258 ],
+    "samples_ts": [ 42.0739, 42.737, 42.0308, 42.8837, 42.2828 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:47:23Z",
+    "avg_ns": 377579804,
+    "stddev_ns": 6351193,
+    "avg_ts": 1356.308108,
+    "stddev_ts": 22.545912,
+    "samples_ns": [ 372424973, 373894780, 373636992, 387431983, 380510292 ],
+    "samples_ts": [ 1374.77, 1369.37, 1370.31, 1321.52, 1345.56 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:47:25Z",
+    "avg_ns": 1305937293,
+    "stddev_ns": 13522740,
+    "avg_ts": 98.022226,
+    "stddev_ts": 1.005768,
+    "samples_ns": [ 1300442432, 1292366252, 1304438971, 1303902694, 1328536116 ],
+    "samples_ts": [ 98.428, 99.0431, 98.1265, 98.1668, 96.3466 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:19:16Z",
+    "avg_ns": 640698554,
+    "stddev_ns": 8442449,
+    "avg_ts": 799.237980,
+    "stddev_ts": 10.456116,
+    "samples_ns": [ 652875468, 635459867, 646229581, 635186003, 633741853 ],
+    "samples_ts": [ 784.223, 805.716, 792.288, 806.063, 807.9 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:19:20Z",
+    "avg_ns": 1748941710,
+    "stddev_ns": 13313898,
+    "avg_ts": 73.190522,
+    "stddev_ts": 0.559300,
+    "samples_ns": [ 1759700788, 1762032558, 1728868153, 1744622542, 1749484513 ],
+    "samples_ts": [ 72.7396, 72.6434, 74.0369, 73.3683, 73.1644 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:21:27Z",
+    "avg_ns": 379833208,
+    "stddev_ns": 2916251,
+    "avg_ts": 1348.023365,
+    "stddev_ts": 10.304461,
+    "samples_ns": [ 377554426, 384109520, 377555613, 381602693, 378343790 ],
+    "samples_ts": [ 1356.1, 1332.95, 1356.09, 1341.71, 1353.27 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:21:29Z",
+    "avg_ns": 1285969483,
+    "stddev_ns": 16587912,
+    "avg_ts": 99.548971,
+    "stddev_ts": 1.276520,
+    "samples_ns": [ 1279748126, 1273538113, 1309616758, 1270516087, 1296428332 ],
+    "samples_ts": [ 100.02, 100.507, 97.7385, 100.746, 98.7328 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:14:01Z",
+    "avg_ns": 2437249430,
+    "stddev_ns": 66433948,
+    "avg_ts": 210.202021,
+    "stddev_ts": 5.922750,
+    "samples_ns": [ 2468232968, 2479407287, 2475612647, 2441614629, 2321379621 ],
+    "samples_ts": [ 207.436, 206.501, 206.817, 209.697, 220.558 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:14:15Z",
+    "avg_ns": 5289126007,
+    "stddev_ns": 63024171,
+    "avg_ts": 24.203349,
+    "stddev_ts": 0.288883,
+    "samples_ns": [ 5363916055, 5306624609, 5325596808, 5238325957, 5211166606 ],
+    "samples_ts": [ 23.8632, 24.1208, 24.0349, 24.4353, 24.5626 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:16:53Z",
+    "avg_ns": 1253292754,
+    "stddev_ns": 79129486,
+    "avg_ts": 409.796891,
+    "stddev_ts": 25.209212,
+    "samples_ns": [ 1190927743, 1190669420, 1208414068, 1318310015, 1358142528 ],
+    "samples_ts": [ 429.917, 430.01, 423.696, 388.376, 376.985 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:17:01Z",
+    "avg_ns": 2909185423,
+    "stddev_ns": 35614966,
+    "avg_ts": 44.003861,
+    "stddev_ts": 0.540189,
+    "samples_ns": [ 2935902159, 2946667991, 2878606593, 2919344964, 2865405409 ],
+    "samples_ts": [ 43.5982, 43.4389, 44.466, 43.8455, 44.6708 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:19:51Z",
+    "avg_ns": 650543418,
+    "stddev_ns": 18555552,
+    "avg_ts": 787.546056,
+    "stddev_ts": 22.428007,
+    "samples_ns": [ 632032527, 670684809, 649115565, 632738336, 668145853 ],
+    "samples_ts": [ 810.085, 763.399, 788.766, 809.181, 766.3 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 16,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:19:55Z",
+    "avg_ns": 1781362880,
+    "stddev_ns": 21643541,
+    "avg_ts": 71.863545,
+    "stddev_ts": 0.869304,
+    "samples_ns": [ 1806353622, 1764741597, 1764660660, 1767373240, 1803685282 ],
+    "samples_ts": [ 70.861, 72.5319, 72.5352, 72.4239, 70.9658 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:21:20Z",
+    "avg_ns": 377608322,
+    "stddev_ns": 6410592,
+    "avg_ts": 1356.211504,
+    "stddev_ts": 22.766437,
+    "samples_ns": [ 372722694, 373942297, 373069875, 387378103, 380928645 ],
+    "samples_ts": [ 1373.68, 1369.2, 1372.4, 1321.71, 1344.08 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 32,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:21:22Z",
+    "avg_ns": 1342539204,
+    "stddev_ns": 21735221,
+    "avg_ts": 95.361608,
+    "stddev_ts": 1.535356,
+    "samples_ns": [ 1368242693, 1364047511, 1323244630, 1330069532, 1327091655 ],
+    "samples_ts": [ 93.5507, 93.8384, 96.7319, 96.2356, 96.4515 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:14:44Z",
+    "avg_ns": 2304834712,
+    "stddev_ns": 9004426,
+    "avg_ts": 222.144447,
+    "stddev_ts": 0.865236,
+    "samples_ns": [ 2319430963, 2305311298, 2304610521, 2298020765, 2296800014 ],
+    "samples_ts": [ 220.744, 222.096, 222.163, 222.8, 222.919 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 4,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:14:58Z",
+    "avg_ns": 5335792102,
+    "stddev_ns": 59840580,
+    "avg_ts": 23.991343,
+    "stddev_ts": 0.267723,
+    "samples_ns": [ 5425666503, 5348961684, 5341956156, 5275207674, 5287168494 ],
+    "samples_ts": [ 23.5916, 23.9299, 23.9613, 24.2644, 24.2096 ]
+  }
+]
@@ -0,0 +1,92 @@
+[
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 512,
+    "n_gen": 0,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:17:31Z",
+    "avg_ns": 1911672545,
+    "stddev_ns": 4048624965,
+    "avg_ts": 363.751611,
+    "stddev_ts": 143.367368,
+    "samples_ns": [ 4771767171, 1201291347, 1195946935, 1197665158, 1191692115 ],
+    "samples_ts": [ 107.298, 426.208, 428.113, 427.498, 429.641 ]
+  },
+  {
+    "build_commit": "e71b80510",
+    "build_number": 10849,
+    "cpu_info": "Intel(R) Xeon(R) Platinum 8480C",
+    "gpu_info": "",
+    "backends": "CPU",
+    "model_filename": "/data/zhoutaichang/.cache/huggingface/hub/models--XHToken--Spark-X2.5-1.7B-GGUF/snapshots/23e1fcac55e7dd71e4c12a23723cc228ba0e5e85/Spark-X2.5-1.7B-Q4_K_M.gguf",
+    "model_type": "spark2_5 1.7B Q4_K - Medium",
+    "model_size": 1102073856,
+    "model_n_params": 1707657216,
+    "n_batch": 2048,
+    "n_ubatch": 512,
+    "n_threads": 8,
+    "cpu_mask": "0x0",
+    "cpu_strict": false,
+    "poll": 50,
+    "type_k": "f16",
+    "type_v": "f16",
+    "n_gpu_layers": -1,
+    "n_cpu_moe": 0,
+    "split_mode": "layer",
+    "main_gpu": 0,
+    "no_kv_offload": false,
+    "flash_attn": -1,
+    "devices": "auto",
+    "tensor_split": "0.00",
+    "tensor_buft_overrides": "none",
+    "load_mode": "auto",
+    "lazy_mode": "auto",
+    "embeddings": false,
+    "no_op_offload": 0,
+    "no_host": false,
+    "fit_target": 0,
+    "fit_min_ctx": 0,
+    "n_prompt": 0,
+    "n_gen": 128,
+    "n_depth": 0,
+    "test_time": "2026-09-11T17:17:46Z",
+    "avg_ns": 2882701640,
+    "stddev_ns": 15288789,
+    "avg_ts": 44.403796,
+    "stddev_ts": 0.236813,
+    "samples_ns": [ 2894272938, 2890318648, 2890951071, 2881201632, 2856763912 ],
+    "samples_ts": [ 44.2253, 44.2858, 44.2761, 44.4259, 44.8059 ]
+  }
+]