From fe3afd93341248f44a2bc009e68d6a2429b53151 Mon Sep 17 00:00:00 2001 From: "Chang Liu (Enterprise Products)" <9713593+chang-l@users.noreply.github.com> Date: Thu, 28 Aug 2025 09:44:28 -0700 Subject: [PATCH 1/2] update doc Signed-off-by: Chang Liu (Enterprise Products) <9713593+chang-l@users.noreply.github.com> --- .../multimodal-feature-support-matrix.md | 22 +++++++++---------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/docs/source/reference/multimodal-feature-support-matrix.md b/docs/source/reference/multimodal-feature-support-matrix.md index bb5175c9da9..af3c195d56b 100644 --- a/docs/source/reference/multimodal-feature-support-matrix.md +++ b/docs/source/reference/multimodal-feature-support-matrix.md @@ -1,13 +1,13 @@ # Multimodal Feature Support Matrix (PyTorch Backend) -| Model | CUDA Graph | Encoder IFB | KV Cache Reuse | Chunked Prefill | -| :----------------- | :--------- | :------------------ | :------------- | :-------------- | -| Gemma 3 | Yes | Yes | No | No | -| HyperCLOVA | Yes | Yes | No | No | -| VILA | Yes | No | No | No | -| LLaVA-NeXT | Yes | Yes | No | No | -| Llama 4 | Yes | No | No | No | -| Mistral-Small-3.1 | Yes | Yes | No | No | -| Phi-4-multimodal | Yes | Yes | No | No | -| Qwen2-VL | Yes | Yes | Yes | No | -| Qwen2.5-VL | Yes | Yes | Yes | No | +| Model Architecture/Feature | Overlap Scheduler | CUDA Graph | Chunked Prefill | Torch Sampler | TLLM C++ Sampler | KV Cache Reuse | Logits Post Processor | EPD Disaggregated Serving | +| ---------------------------------- | ----------------- | ---------- | --------------- | ------------- | ---------------- | -------------- | --------------------- | ------------------------- | +| Gemma3ForConditionalGeneration | Yes | Yes | N/A | Yes | Yes | N/A | Yes | No | +| HCXVisionForCausalLM | Yes | Yes | No | Yes | Yes | No | Yes | No | +| LlavaLlamaModel (VILA) | Yes | Yes | No | Yes | Yes | No | Yes | No | +| LlavaNextForConditionalGeneration | Yes | Yes | No | Yes | Yes | Yes | Yes | No | +| Llama4ForConditionalGeneration | Yes | Yes | No | Yes | Yes | No | Yes | No | +| Mistral3ForConditionalGeneration | Yes | Yes | No | Yes | Yes | No | Yes | No | +| Phi4MMForCausalLM | Yes | Yes | No | Yes | Yes | No | Yes | No | +| Qwen2VLForConditionalGeneration | Yes | Yes | No | Yes | Yes | Yes | Yes | No | +| Qwen2_5_VLForConditionalGeneration | Yes | Yes | No | Yes | Yes | Yes | Yes | No | From 65f822d818234d516980dd6693bf4e26e1e16562 Mon Sep 17 00:00:00 2001 From: "Chang Liu (Enterprise Products)" <9713593+chang-l@users.noreply.github.com> Date: Thu, 28 Aug 2025 09:59:16 -0700 Subject: [PATCH 2/2] update Signed-off-by: Chang Liu (Enterprise Products) <9713593+chang-l@users.noreply.github.com> --- docs/source/reference/multimodal-feature-support-matrix.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/source/reference/multimodal-feature-support-matrix.md b/docs/source/reference/multimodal-feature-support-matrix.md index af3c195d56b..ed6db116f31 100644 --- a/docs/source/reference/multimodal-feature-support-matrix.md +++ b/docs/source/reference/multimodal-feature-support-matrix.md @@ -5,7 +5,7 @@ | Gemma3ForConditionalGeneration | Yes | Yes | N/A | Yes | Yes | N/A | Yes | No | | HCXVisionForCausalLM | Yes | Yes | No | Yes | Yes | No | Yes | No | | LlavaLlamaModel (VILA) | Yes | Yes | No | Yes | Yes | No | Yes | No | -| LlavaNextForConditionalGeneration | Yes | Yes | No | Yes | Yes | Yes | Yes | No | +| LlavaNextForConditionalGeneration | Yes | Yes | No | Yes | Yes | No | Yes | No | | Llama4ForConditionalGeneration | Yes | Yes | No | Yes | Yes | No | Yes | No | | Mistral3ForConditionalGeneration | Yes | Yes | No | Yes | Yes | No | Yes | No | | Phi4MMForCausalLM | Yes | Yes | No | Yes | Yes | No | Yes | No |