tzhouamandClaude Opus 5
9264ff81ab
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
35 lines
1.5 KiB
Plaintext
35 lines
1.5 KiB
Plaintext
vllm-embedding
|
|
Copyright 2026 tzhouam
|
|
|
|
This product is a derivative work of vLLM-Omni
|
|
(https://github.com/vllm-project/vllm-omni), licensed under the Apache License,
|
|
Version 2.0. Upstream code is Copyright the vLLM-Omni contributors, and vLLM-Omni
|
|
is itself a derivative of vLLM (https://github.com/vllm-project/vllm).
|
|
|
|
Modifications in this fork are confined to edge/CPU inference and are listed in
|
|
README.md; the substantive ones are:
|
|
|
|
vllm_omni/model_executor/layers/quantization/cpu_int4.py
|
|
4-bit weight-only quantization for the CPU backend
|
|
vllm_omni/model_executor/models/spark2_5/spark2_5.py
|
|
RoPE table sized to max_model_len rather than the advertised context
|
|
vllm_omni/edge/
|
|
device-class weight-path selection, KV budget sizing, memory release
|
|
vllm_omni/platforms/cpu/, vllm_omni/patch.py
|
|
CPU platform and CPU-backend patches
|
|
benchmarks/edge_harness/
|
|
measurement tooling and results (original to this fork)
|
|
baselines/, docs/engines/, docs/analysis/
|
|
benchmark results and written analyses (original to this fork)
|
|
|
|
vLLM itself is not modified and no vLLM source is redistributed here. This fork
|
|
changes vLLM's CPU behaviour through its own extension points at runtime
|
|
(vllm_omni/patch.py); the pinned upstream versions are recorded in
|
|
docs/edge/environment.md.
|
|
|
|
Baseline measurements reference llama.cpp (https://github.com/ggml-org/llama.cpp,
|
|
MIT). No llama.cpp source is redistributed; only benchmark output produced with
|
|
it, under baselines/llama_cpp/.
|
|
|
|
Upstream's README is preserved at docs/README.vllm-omni.md.
|