@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 5bbaf492a6
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 14, 22:35 UTC

[Doc] Update partial support (#21916)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

Cyrus Leung committed Jul 30, 2025 at 16:32 UTC 5bbaf492a6238ff517249e73151ae9989f7bea9e
1 file changed +4 -3
docs/features/compatibility_matrix.md
+4 -3
@@ -41,17 +41,18 @@ th:not(:first-child) {
41 | [LoRA](lora.md) | ✅ | ✅ | ✅ | | | | | | | | | | | |
42 | [SD](spec_decode.md) | ✅ | ✅ | ❌ | ✅ | | | | | | | | | | |
43 | CUDA graph | ✅ | ✅ | ✅ | ✅ | ✅ | | | | | | | | | |
44 -| [pooling](../models/pooling_models.md) | ✅\* | ✅\* | ✅ | ❌ | ✅ | ✅ | | | | | | | | |
44 +| [pooling](../models/pooling_models.md) | 🟠\* | 🟠\* | ✅ | ❌ | ✅ | ✅ | | | | | | | | |
45 | <abbr title="Encoder-Decoder Models">enc-dec</abbr> | ❌ | [❌](gh-issue:7366) | ❌ | [❌](gh-issue:7366) | ✅ | ✅ | ✅ | | | | | | | |
46 | <abbr title="Logprobs">logP</abbr> | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | | | | | | |
47 | <abbr title="Prompt Logprobs">prmpt logP</abbr> | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | | | | | |
48 | <abbr title="Async Output Processing">async output</abbr> | ✅ | ✅ | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | | | | |
49 | multi-step | ❌ | ✅ | ❌ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | | | |
50 -| [mm](multimodal_inputs.md) | ✅ | ✅ | [🟠](gh-pr:4194) | ❔ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❔ | ✅ | | |
50 +| [mm](multimodal_inputs.md) | ✅ | ✅ | [🟠](gh-pr:4194)<sup>^</sup> | ❔ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❔ | ✅ | | |
51 | best-of | ✅ | ✅ | ✅ | [❌](gh-issue:6137) | ✅ | ❌ | ✅ | ✅ | ✅ | ❔ | [❌](gh-issue:7968) | ✅ | ✅ | |
52 | beam-search | ✅ | ✅ | ✅ | [❌](gh-issue:6137) | ✅ | ❌ | ✅ | ✅ | ✅ | ❔ | [❌](gh-issue:7968) | ❔ | ✅ | ✅ |
53
54 -\* Chunked prefill and prefix caching are only applicable to last-token pooling.
54 +\* Chunked prefill and prefix caching are only applicable to last-token pooling.
55 +<sup>^</sup> LoRA is only applicable to the language backbone of multimodal models.
56
57 [](){ #feature-x-hardware }
58