Officially support SmolLM3 using the Transformers backend (#22665)

Signed-off-by: Harry Mellor <[email protected]>

Author: hmellor

docs/models/supported_models.md

@@ -409,6 +409,12 @@ th {
 | `MiniMaxText01ForCausalLM` | MiniMax-Text | `MiniMaxAI/MiniMax-Text-01`, etc. | | | ✅︎ |
 | `Zamba2ForCausalLM` | Zamba2 | `Zyphra/Zamba2-7B-instruct`, `Zyphra/Zamba2-2.7B-instruct`, `Zyphra/Zamba2-1.2B-instruct`, etc. | | | ✅︎ |
 
+Some models are supported only via the [Transformers backend](#transformers). The purpose of the table below is to acknowledge models which we officially support in this way. The logs will say that the Transformers backend is being used, and you will see no warning that this is fallback behaviour. This means that, if you have issues with any of the models listed below, please [make an issue](https://github.com/vllm-project/vllm/issues/new/choose) and we'll do our best to fix it!
+
+| Architecture | Models | Example HF Models | [LoRA](../features/lora.md) | [PP](../serving/parallelism_scaling.md) | [V1](gh-issue:8779) |
+|--------------|--------|-------------------|----------------------|---------------------------|---------------------|
+| `SmolLM3ForCausalLM` | SmolLM3 | `HuggingFaceTB/SmolLM3-3B` | ✅︎ | ✅︎ | ✅︎ |
+
 !!! note
     Currently, the ROCm version of vLLM supports Mistral and Mixtral only for context lengths up to 4096.
 

tests/models/registry.py

@@ -291,6 +291,7 @@ def check_available_online(
     "Qwen3ForCausalLM": _HfExamplesInfo("Qwen/Qwen3-8B"),
     "Qwen3MoeForCausalLM": _HfExamplesInfo("Qwen/Qwen3-30B-A3B"),
     "RWForCausalLM": _HfExamplesInfo("tiiuae/falcon-40b"),
+    "SmolLM3ForCausalLM": _HfExamplesInfo("HuggingFaceTB/SmolLM3-3B"),
     "StableLMEpochForCausalLM": _HfExamplesInfo("stabilityai/stablelm-zephyr-3b"),  # noqa: E501
     "StableLmForCausalLM": _HfExamplesInfo("stabilityai/stablelm-3b-4e1t"),
     "Starcoder2ForCausalLM": _HfExamplesInfo("bigcode/starcoder2-3b"),

vllm/model_executor/models/registry.py

@@ -271,6 +271,9 @@
 }
 
 _TRANSFORMERS_SUPPORTED_MODELS = {
+    # Text generation models
+    "SmolLM3ForCausalLM": ("transformers", "TransformersForCausalLM"),
+    # Multimodal models
     "Emu3ForConditionalGeneration": ("transformers", "TransformersForMultimodalLM"),  # noqa: E501
 }