Morph MORPH
SIGN IN SIGN UP
dhruvhead / recipes UNCLAIMED
0 0 3 JavaScript

COMMITS

main
July 8, 2026
J
Hy3 rocm verified (#620)
jotsaiamd committed
ecdf173
C
[New Model] Add MOSS-Transcribe-Diarize recipes (#622)
Canlin Guo committed
7a40784
July 7, 2026
Y
fix h20*8 deepseek-v4-pro-dspark bug (#617)
yiminghub2024 committed
37fc085
H
[AMD][ROCm]Minimax-M3 Eagle3 MTP attn backend override for perf (#615)
Hongxia Yang committed
2aa7621
July 6, 2026
H
[ROCm][AMD] MiniMax-M3 MXFP8 MI355x recipe update (#581)
Hongxia Yang committed
1ad1537
T
Pin Hy3 AMD docker image to vllm/vllm-openai-rocm:latest (#613)
Tiezhen WANG committed
80cb320
T
Add tencent/Hy3 recipe (#612)
Tiezhen WANG committed
3208243
J
Add quantised variants to Laguna XS.2 and XS-2.1 recipes (#610)
Joe Rowell committed
2610112
July 2, 2026
J
Add Laguna-XS-2.1 recipe (#608)
Joe Rowell committed
ba26d6e
R
Add single-select spec_decoding modes with variant/mode coupling (#607)
Roy Wang committed
28ac1b3
R
Add NVFP4 variant to Qwen3.6-27B recipe (#606)
Roy Wang committed
3328d47
June 28, 2026
R
Update prompt guide for Unlimited OCR (#588)
Roger Wang committed
114b508
J
Unlimited-OCR recipe (#587)
Jee Jee Li committed
451d842
June 27, 2026
R
Add NVFP4 variant to GLM-5.2 recipe (#586)
Roy Wang committed
bad7e52
June 26, 2026
M
SEO: canonicalize on recipes.vllm.ai + make model pages discoverable (#584)
Michael Goin committed
710c850
June 25, 2026
Y
LFM2.5: fix context_length to 32K (per model cards) + Blackwell oink + copy cleanup (#578)
Yi Zhong committed
fe81a78
F
[ROCm] Add MI355X-only MiniMax-M3 MXFP4 variant (#580)
functionstackx committed
e2f883e
R
Add LiquidAI/LFM2.5-230M recipe (#576)
Roy Wang committed
05d2eaa
Y
Add LiquidAI LFM2.5 recipes (dense, MoE, VL) (#575)
Yi Zhong committed
598154c
June 24, 2026
R
Update max sequence number from 2 to 16 in strategy overrides for Dee… (#573)
Roy Wang committed
da310e6
J
Update context length for Laguna-XS.2 to 256K (#564)
Joe Rowell committed
63aceeb
June 23, 2026
L
add Xeon 6 CPUs support for Meta models (#518)
Louie Tsai committed
0a07614
F
Update qwen3.6 serving config (#536)
Faradawn Yang committed
e8e8b45
June 18, 2026
T
[ROCm] Add MI300X and MI355X guidance for Moonshot Kimi (#571)
Tan Pin Siang committed
aecd2f5
A
Fix `--max-num-seqs 512` for `Qwen/Qwen3.6-27B-FP8` (#551)
Alvaro Bartolome committed
2b82044
R
Add poolside/Laguna-M.1 recipe (#569)
Roy Wang committed
80363de
W
[Bug] Fix spec config in recipe (#563)
Wentao Ye committed
70bd67e
T
GLM-5.2: use fp8 KV cache on Hopper (Blackwell/AMD keep fp8_e4m3) (#566)
Tiezhen WANG committed
6d0eae5
L
add Xeon 6 support for Qwen models (#519)
Louie Tsai committed
490e06b
June 17, 2026
Y
Add vLLM-Omni TTS recipes (12 models across 8 providers) (#554)
Yueqian Lin committed
c8dd60d