Hongxia Yangandtjtanaa e28e8c8782 [ROCm][Quant] Minimax-M3: Enable fp8_per_channel for bf16 weights on mi300x (#45854) ...

Signed-off-by: Hongxia Yang <hongxia.yang@amd.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>

2026-06-17 12:02:40 +00:00

..

assets

Add vLLM library info to Hugging Face Hub requests (#43857)

2026-05-29 14:04:58 +00:00

benchmarks

[Model] Add DiffusionGemma Support (#45163)

2026-06-11 22:17:35 -07:00

compilation

[Bugfix][V1] Clean up compiled-model bytecode hooks on VllmRunner exit (#45195)

2026-06-16 20:31:17 -07:00

config

[ModelRunnerV2] Various model/config compatibility fixes (#45868)

2026-06-17 03:23:01 +00:00

cute_utils

[GDN] GDN Prefill kernel for SM100 (#43273)

2026-05-26 14:02:11 +08:00

device_allocator

[XPU][Feature] transparent sleep mode support for XPU platform (#37149)

2026-06-07 10:45:31 +08:00

distributed

[KV Connector][Offloading] Avoid blocking the engine to flush offloads on idle (#45595)

2026-06-17 11:35:07 +03:00

engine

[Core] Add prefill step cadence for better non-PD DP balancing (#44558)

2026-06-16 13:17:18 -07:00

entrypoints

[Rust Frontend] Support hybrid/external DP LB in Python supervised bootstrap (#45805)

2026-06-17 07:32:40 +00:00

inputs

feat: Enable prompt_embeds Content Part Support in vLLM Chat Completions API (#40720)

2026-05-01 10:05:55 +08:00

ir

Speed up docs build (#44635)

2026-06-05 14:51:44 +00:00

kernels

[Kernel] Add weightless RMSNorm CUDA kernels for has_weight=False (#41430) (#44109)

2026-06-16 23:45:55 -07:00

logging_utils

[vLLM IR] 1/N Implement IR skeleton and rms_norm op (#33825)

2026-03-31 22:15:05 -04:00

lora

[Bugfix][MoE] Fix fused MoE expert mapping helper call sites (#44897)

2026-06-08 13:35:04 -07:00

model_executor

[ROCm][Quant] Minimax-M3: Enable fp8_per_channel for bf16 weights on mi300x (#45854)

2026-06-17 12:02:40 +00:00

models

[M3] Enable FP8 sparse GQA (#45744)

2026-06-16 21:38:03 -07:00

multimodal

fix(security): enforce audio decode duration limit in chat completions path (#45908)

2026-06-17 11:07:13 +00:00

parser

[Bugfix][Gemma4] Pre-initialise streaming reasoning state when prompt ends inside an open <|channel> (fixes #45834) (#45852)

2026-06-17 06:16:02 -04:00

platforms

[ROCm][Quant] Minimax-M3: Enable fp8_per_channel for bf16 weights on mi300x (#45854)

2026-06-17 12:02:40 +00:00

plugins

Add vLLM library info to Hugging Face Hub requests (#43857)

2026-05-29 14:04:58 +00:00

profiler

[Refactor] Clean up log once scope="local" (#40540)

2026-04-22 16:42:43 -04:00

ray

…

reasoning

[Bugfix][Gemma4] Pre-initialise streaming reasoning state when prompt ends inside an open <|channel> (fixes #45834) (#45852)

2026-06-17 06:16:02 -04:00

renderers

[Frontend] Remove AsyncMicrobatchTokenizer. (#45759)

2026-06-16 08:02:11 +00:00

third_party

…

tokenizers

Register parsed config classes before tokenizer init (#40299)

2026-06-16 05:33:08 +00:00

tool_parsers

[Bugfix] Gemma4: skip forced JSON for required/named tool choice (#45795)

2026-06-16 20:38:29 +00:00

tracing

…

transformers_utils

Register parsed config classes before tokenizer init (#40299)

2026-06-16 05:33:08 +00:00

triton_utils

[Perf] Add VLLM_TRITON_FORCE_FIRST_CONFIG to skip Triton autotuning (#42425)

2026-06-16 15:16:45 +02:00

usage

replace cuda_device_count_stateless() to current_platform.device_count() (#37841)

2026-03-31 22:32:54 +08:00

utils

[Bugfix][Core] Fall back when numactl --membind is blocked in constrained containers (#45438)

2026-06-16 09:49:56 -07:00

v1

[ROCm][Bugfix]: Fallback GFX942 sparse MLA ops to Triton (#45782)

2026-06-17 11:41:29 +00:00

vllm_flash_attn

feat: MLA prefill enable FA4 fp8 output (#43050)

2026-06-16 07:10:59 -07:00

__init__.py

…

_aiter_ops.py

[ROCm][gpt-oss] Pass GateMode.INTERLEAVE for MXFP4 W4A16 fused MoE (#44893)

2026-06-12 01:02:04 -05:00

_custom_ops.py

[Kernel] Add weightless RMSNorm CUDA kernels for has_weight=False (#41430) (#44109)

2026-06-16 23:45:55 -07:00

_xpu_ops.py

[BUGFIX][XPU] Update fa interface for compatibility (#45394)

2026-06-12 11:45:01 +00:00

collect_env.py

[XPU]Enhance environment collection for Intel XPU and optimize layout (#35698)

2026-04-13 12:51:46 +08:00

connections.py

[CI] Add retry with 4x backoff to HTTP fetches for transient failures (#37218)

2026-03-19 19:00:20 +00:00

env_override.py

[Perf] Add VLLM_TRITON_FORCE_FIRST_CONFIG to skip Triton autotuning (#42425)

2026-06-16 15:16:45 +02:00

envs.py

[Perf] Add VLLM_TRITON_FORCE_FIRST_CONFIG to skip Triton autotuning (#42425)

2026-06-16 15:16:45 +02:00

exceptions.py

[openapi server] log exception in exception handler(2/N) (#36201)

2026-03-10 20:16:30 -07:00

forward_context.py

[Perf] Set IR Op Priority Once at Worker Init (#42631)

2026-05-15 15:56:13 +00:00

logger.py

[Log] Log once in local node by default (#37568)

2026-03-19 12:04:59 -07:00

logits_process.py

…

logprobs.py

…

model_inspection.py

…

outputs.py

Revert "[Core] Replace routing replay with device cache and async D2H pipeline" (#39917) (#42434)

2026-05-13 23:49:01 -07:00

pooling_params.py

[Misc] Replace assert with proper exceptions for security and validation in pooling (#43286)

2026-05-22 18:43:33 +08:00

py.typed

…

sampling_params.py

[Bugfix] Reject out-of-range temperature values in SamplingParams (#44965)

2026-06-14 23:12:44 -07:00

scalar_type.py

…

scripts.py

…

sequence.py

[Core] Cleanup KVConnector handling with PP + fix MRV2 (#43732)

2026-05-28 13:12:03 -07:00

tasks.py

[Frontend] Remove frontend pooling multi task support. (#37861)

2026-04-21 12:27:44 +00:00

version.py

…