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, 05:30 UTC
[Docs] Update V1 doc to reflect whisper support (#24606)
Signed-off-by: Russell Bryant <rbryant@redhat.com>
Russell Bryant committed
Sep 11, 2025 at 04:50 UTC
ba6011027dd68ce3bff9de9d8001354f095cee5c
2 files changed
+5
-4
docs/models/supported_models.md
+1
-1
@@ -766,7 +766,7 @@ Speech2Text models trained specifically for Automatic Speech Recognition.
766
767
| Architecture | Models | Example HF Models | [LoRA](../features/lora.md) | [PP](../serving/parallelism_scaling.md) | [V1](gh-issue:8779) |
768
|--------------|--------|-------------------|----------------------|---------------------------|---------------------|
769
-| `WhisperForConditionalGeneration` | Whisper | `openai/whisper-small`, `openai/whisper-large-v3-turbo`, etc. | | | |
769
+| `WhisperForConditionalGeneration` | Whisper | `openai/whisper-small`, `openai/whisper-large-v3-turbo`, etc. | | | ✅︎ |
770
| `VoxtralForConditionalGeneration` | Voxtral (Mistral format) | `mistralai/Voxtral-Mini-3B-2507`, `mistralai/Voxtral-Small-24B-2507`, etc. | ✅︎ | ✅︎ | ✅︎ |
771
| `Gemma3nForConditionalGeneration` | Gemma3n | `google/gemma-3n-E2B-it`, `google/gemma-3n-E4B-it`, etc. | | | ✅︎ |
772
docs/usage/v1_guide.md
+4
-3
@@ -83,7 +83,7 @@ based on assigned priority, with FCFS as a tie-breaker), configurable via the
83
| Model Type | Status |
84
|-----------------------------|------------------------------------------------------------------------------------|
85
| **Decoder-only Models** | <nobr>🚀 Optimized</nobr> |
86
-| **Encoder-Decoder Models** | <nobr>🟠 Delayed</nobr> |
86
+| **Encoder-Decoder Models** | <nobr>🟢 Whisper only</nobr> |
87
| **Embedding Models** | <nobr>🟢 Functional</nobr> |
88
| **Mamba Models** | <nobr>🟢 (Mamba-2), 🟢 (Mamba-1)</nobr> |
89
| **Multimodal Models** | <nobr>🟢 Functional</nobr> |
@@ -118,8 +118,9 @@ Please note that prefix caching is not yet supported for any of the above models
118
119
#### Encoder-Decoder Models
120
121
-Models requiring cross-attention between separate encoder and decoder (e.g., `BartForConditionalGeneration`, `MllamaForConditionalGeneration`)
122
-are not yet supported.
121
+Whisper is supported. Other models requiring cross-attention between separate
122
+encoder and decoder (e.g., `BartForConditionalGeneration`,
123
+`MllamaForConditionalGeneration`) are not yet supported.
124
125
### Features
126