Morph MORPH
SIGN IN SIGN UP
dhruvhead / aphrodite-engine UNCLAIMED
0 0 0 C++

COMMITS

main
March 12, 2026
A
feat: add support for the Qwen3.5 family of models (#1624)
AlpinDale committed
1eab18a
February 17, 2026
L
fix: compute engine max_concurrency from worker KV cache configs to match runtime reporting (#1622)
lucy committed
a63d017
January 21, 2026
A
fix: mark GLM-4 MoE Lite as an MLA model (#1621)
AlpinDale committed
f0f652a
January 19, 2026
A
[models] add support for GLM-4.7 Flash (#1620)
AlpinDale committed
79a8216
January 7, 2026
A
fix: tokenizer server init (#1617)
AlpinDale committed
212cb10
November 22, 2025
A
[cli] add CLI arg for selecting attention backend (#1612)
AlpinDale committed
82f869f
November 19, 2025
A
[fix] log message killing compilation
AlpinDale committed
19dddde
November 16, 2025
A
[logger][metrics] log number of cache hits in the request-level logger (#1611)
AlpinDale committed
1aef37b
A
[cli][diffusion] only import diffusion backend when it is called (#1610)
AlpinDale committed
163bfb2
November 10, 2025
A
[diffusion] `aphrodite diffusion` backend (#1607)
AlpinDale committed
cb753df
A
[engine] add API for concurrency rate and kv cache token limit (#1608)
AlpinDale committed
56ea41d
November 8, 2025
A
[readme] update installation guide in readme
AlpinDale committed
5b12d09
A
[build] fix aphrodite-kernels wheel installation for pypi compat (#1606)
AlpinDale committed
5e3afa0
A
[build][kernels] isolated aphrodite kernel library (#1602)
AlpinDale committed
3cce659
November 6, 2025
A
[multi node] better cluster example script (#1605)
AlpinDale committed
c9daca2
5
[structured outputs] require structured output parameters to be explicitly None or valid (#1604)
50h100a committed
dfec7ed
November 5, 2025
A
[build] upgrade flashinfer to 0.5.1 (#1601)
AlpinDale committed
a9664b0
A
[quant] fix GLM-4.5V AWQ (#1600)
AlpinDale committed
8b9df85
A
[lora][moe] fix MoE models by registering the correct op (#1599)
AlpinDale committed
913d0c3
November 4, 2025
A
[build] downgrade flashinfer to 0.4.1 (#1598)
AlpinDale committed
eb82a1c
A
[sync] sync to upstream 03c4c4a (#1597)
AlpinDale committed
5495210
A
[offloader] fix async scheduling support with KV cache offloader (#1596)
AlpinDale committed
7016e79
A
[api] bring back anthropic /v1/messages endpoint in OpenAI server (#1595)
AlpinDale committed
cd631de
A
[sampler] fix mixed penalties in batch with async scheduling (#1594)
AlpinDale committed
b5d3e11
A
[TPU] prevent single-process DP (#1593)
AlpinDale committed
5bce45e
A
[python3.10] import `Self` from `typing_extensions` (#1592)
AlpinDale committed
2c3b58c
A
[spec] fix DeepSeek v3.2 MTP metadata and cuda graph (#1591)
AlpinDale committed
7fd7f4a
A
[v0] remove `APHRODITE_USE_V1` from platform and v1 (#1590)
AlpinDale committed
e5538f1
A
[kvoffload] feat: make LMCache connecter work (#1589)
AlpinDale committed
52d12ec
A
[fix] engine args import
AlpinDale committed
5c39abb