anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity
3,094 Commits 235 Branches 31 Tags
5d81c3678b5f7dba0888b2d34315212f18a8a054
Commit Graph
864 Commits
Author SHA1 Message Date
tzhouam 5d81c3678b Audit Qualcomm components and Windows TTS sustained drift 2026-09-24 09:12:11 +08:00
tzhouam d8995de9ec Record official InternVLA dataset access gate 2026-09-24 08:49:40 +08:00
tzhouam bc168a3ee2 Audit SA talker CPU and NPU vocoder failure 2026-09-24 08:43:27 +08:00
tzhouam 6af714cf9b Track SA8775P InternVLA Cosmos encoder artifact 2026-09-24 08:37:15 +08:00
tzhouam 11c1a56467 Profile mobile MiniCPM and InternVLA NPU components 2026-09-24 08:32:04 +08:00
tzhouam 3b4469a136 Record X Elite CPU talker parity and RB3 fallback 2026-09-24 08:26:45 +08:00
tzhouam 855dd31f07 Track RB3 speech fallback and X Elite Cosmos export 2026-09-24 08:23:03 +08:00
tzhouam afd3f8c716 Record X Elite MiniCPM speech head inference 2026-09-24 08:18:41 +08:00
tzhouam e5aaf3186c Record SA speech head and X Elite QNN vocoder results 2026-09-24 08:16:15 +08:00
tzhouam 76941854b1 Record SA8775P QNN vocoder compile and inference 2026-09-24 08:11:27 +08:00
tzhouam 4d6124f320 Record SA8775P CPU vocoder placement and timing 2026-09-24 08:06:23 +08:00
tzhouam f04fdf79da Track CPU talker controls and mobile speech-head profiles 2026-09-24 08:01:05 +08:00
tzhouam d71d39f656 Audit mobile speech heads and S25 VLA latent impact 2026-09-24 07:56:17 +08:00
tzhouam 1949c31f80 Advance Qualcomm TTS and InternVLA component qualification 2026-09-24 07:32:21 +08:00
tzhouam ea373e3cbf Audit X Elite CPU vocoder and isolate TFLite conversion parity 2026-09-24 07:16:15 +08:00
tzhouam e4a4477929 Update device-model matrix with Qualcomm TTS component evidence 2026-09-24 07:04:24 +08:00
tzhouam 152d67ab86 Audit Qualcomm Qwen3-TTS vocoder parity and profile 2026-09-24 06:48:14 +08:00
tzhouam bd39e903e1 Profile Spark attention across Qualcomm edge devices 2026-09-24 06:11:15 +08:00
tzhouam c5f47b22fd Profile repeated Radeon MiniCPM-o requests and record memory regression 2026-09-24 03:09:21 +08:00
tzhouam 5dd5d70671 Profile native Windows Radeon InternVLA policy stage 2026-09-24 02:24:16 +08:00
tzhouam 752357d318 Profile native Windows CPU InternVLA through Omni 2026-09-24 02:05:26 +08:00
tzhouam 48646da78e Add admitted InternVLA policy stages and update device matrix 2026-09-24 01:42:41 +08:00
tzhouam 9df5ac8836 Run InternVLA synthetic policy with Radeon Cosmos handoff 2026-09-24 01:12:27 +08:00
tzhouam b796c787ec Profile InternVLA Cosmos encoder on Radeon 890M and update matrix 2026-09-24 00:51:17 +08:00
tzhouam e6ea35ca12 Qualify scoped WSL CPU Qwen3-TTS Omni complete requests 2026-09-24 00:41:14 +08:00
tzhouam 10ae2818c7 Profile MiniCPM-o GGUF Omni complete path on WSL CPU 2026-09-24 00:21:11 +08:00
tzhouam fa5aadca6f Record MiniCPM-o GGUF Omni desktop qualification depth 2026-09-23 23:02:10 +08:00
tzhouam 13cc723c71 Qualify scoped MiniCPM-o Radeon hybrid path 2026-09-23 22:43:20 +08:00
tzhouam 445ca92ef1 Record MiniCPM-o CPU GGUF complete-path evidence 2026-09-23 22:22:54 +08:00
tzhouam bb8735effc Add native Windows CPU Qwen TTS Omni stage and profile evidence 2026-09-23 21:18:21 +08:00
tzhouam 8569995591 Add bounded Omni Qwen3-TTS CPU Radeon whole-session stage 2026-09-23 20:52:43 +08:00
tzhouam 34beb7f01e Add bounded Omni llama.cpp Spark text stage for Windows CPU and Radeon 2026-09-23 20:30:28 +08:00
tzhouam 4fee1dbe39 Profile Qwen3-TTS hybrid Radeon and CPU route 2026-09-23 19:59:04 +08:00
tzhouam 54c086acae Record native Windows CPU Qwen3-TTS scoped E2E evidence 2026-09-23 19:19:13 +08:00
tzhouam 03d6cb9a93 Qualify standalone Spark text on Windows CPU and Radeon 890M 2026-09-23 18:57:41 +08:00
tzhouam 34295f3eb3 Qualify Spark output-head component on AMD NPU and update E2E matrix 2026-09-23 18:32:13 +08:00
tzhouam 10221f0665 Qualify scoped MiniCPM-o image and audio paths on WSL CPU 2026-09-23 17:44:02 +08:00
tzhouam b29a6ca99f Document Windows MiniCPM-o image and audio end-to-end runs 2026-09-23 17:29:13 +08:00
tzhouam a8fdb1babf Validate MiniCPM-o audio input to speech on WSL GPU 2026-09-23 17:09:37 +08:00
tzhouam 0d00ba4701 Validate MiniCPM-o image to speech on WSL GPU 2026-09-23 17:03:31 +08:00
tzhouam 8a4559c46c Validate MiniCPM-o speech pipeline on native Windows GPU 2026-09-23 16:54:17 +08:00
tzhouam 2f5d0edcf7 Qualify scoped MiniCPM-o CPU speech and Windows InternVLA policy paths 2026-09-23 16:21:28 +08:00
tzhouam c0deaa76ba Repair MiniCPM-o and InternVLA scoped E2E paths 2026-09-23 15:15:07 +08:00
tzhouam b039b6a5bd Enable scoped Qwen3.8 Omni inference and CPU TTS compatibility 2026-09-23 10:01:21 +08:00
tzhouam 75b65f859b Add local E2E profiling evidence and support roadmap 2026-09-22 19:25:02 +08:00
tzhouam ab2f2a3d65 Audit all mobile and PC model pairings with archived evidence 2026-09-22 13:54:34 +08:00
tzhouam 1887d8294b Present support as mobile and PC by model matrix 2026-09-22 13:31:08 +08:00
tzhouam 5c75d5bb0a Document support as model and PC configuration pairings 2026-09-22 13:20:50 +08:00
tzhouam 723c550209 Integrate portable edge graph backends with Omni runtime 
Extract versioned stage contracts, model shared CPU/iGPU/NPU memory constraints, and add explicit graph routes, reservations, output acknowledgements, and worker retirement. Fix native Windows shared-buffer lifetime and host paths. Document qualified hardware paths and remaining rollout gates.
 2026-09-22 13:12:33 +08:00
tzhouamandClaude Fable 5.1 33de5d8629 Merge vLLM-Omni v0.29.0rc1 into edge-infer/omni-0.29.0 
The pair for vLLM v0.29.0: rc1's docs pin vllm==0.29.0 and a static scan of
rc1's vllm.* imports against the v0.29.0 tree finds nothing missing, while
our 0.28-based tree needs seven modules v0.29.0 no longer has
(vllm.inputs.preprocess, entrypoints.launcher, openai.cli_args,
openai.engine.protocol, serve.utils.server_utils, layers.batch_invariant,
models.hunyuan_v1).

Ten files conflicted. Upstream's structure is the base everywhere; the
fork's features are re-attached to it:
- deploy_profile / apply_deploy_profile run before rc1's resolve_omni_config
  pops deploy_config (async_omni_engine, cli/serve); the deploy-profile
  selector and the resolve_model_config_path chain it needs stay in
  entrypoints/utils.py; stage_overrides takes rc1's dict type.
- config_factory keeps the lazy diffusion output-type import; the diffusion
  registry import in rc1's new config/resolver.py is made lazy for the same
  WP6 reason (TTS path must not load diffusion).
- requirements: our core/media split kept; rc1's transformers>=5.13 floor
  and kernels==0.16.1 applied in core.txt / media.txt.
- TTS: rc1's split_granularity and bounded seed plus our output_mode=
  codec_tokens, _SessionInbox cancel handling and codec DELTA accumulation;
  our private _conditioning_cache_salt is superseded by rc1's public
  tts_adapters/base.py::conditioning_cache_salt.
- stage_runtime: rc1's single ExitStack G2 path (lock released before READY)
  replaces our two-path block, with the four worker_phase names the init
  profiler consumes still emitted; _cpu_placed / host admission kept.

Not yet run against vLLM 0.29.0: the tests listed in the W6 record.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-09-18 09:57:20 +08:00