TPUΒΆ
Supported ModelsΒΆ
Text-only Language ModelsΒΆ
Model | Architecture | Supported |
---|---|---|
mistralai/Mixtral-8x7B-Instruct-v0.1 | MixtralForCausalLM | π¨ |
mistralai/Mistral-Small-24B-Instruct-2501 | MistralForCausalLM | β |
mistralai/Codestral-22B-v0.1 | MistralForCausalLM | β |
mistralai/Mixtral-8x22B-Instruct-v0.1 | MixtralForCausalLM | β |
meta-llama/Llama-3.3-70B-Instruct | LlamaForCausalLM | β |
meta-llama/Llama-3.1-8B-Instruct | LlamaForCausalLM | β |
meta-llama/Llama-3.1-70B-Instruct | LlamaForCausalLM | β |
meta-llama/Llama-4-* | Llama4ForConditionalGeneration | β |
microsoft/Phi-3-mini-128k-instruct | Phi3ForCausalLM | π¨ |
microsoft/phi-4 | Phi3ForCausalLM | β |
google/gemma-3-27b-it | Gemma3ForConditionalGeneration | π¨ |
google/gemma-3-4b-it | Gemma3ForConditionalGeneration | β |
deepseek-ai/DeepSeek-R1 | DeepseekV3ForCausalLM | β |
deepseek-ai/DeepSeek-V3 | DeepseekV3ForCausalLM | β |
RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8 | LlamaForCausalLM | β |
RedHatAI/Meta-Llama-3.1-70B-Instruct-quantized.w8a8 | LlamaForCausalLM | β |
Qwen/Qwen3-8B | Qwen3ForCausalLM | β |
Qwen/Qwen3-32B | Qwen3ForCausalLM | β |
Qwen/Qwen2.5-7B-Instruct | Qwen2ForCausalLM | β |
Qwen/Qwen2.5-32B | Qwen2ForCausalLM | β |
Qwen/Qwen2.5-14B-Instruct | Qwen2ForCausalLM | β |
Qwen/Qwen2.5-1.5B-Instruct | Qwen2ForCausalLM | π¨ |
β
Runs and optimized.
π¨ Runs and correct but not optimized to green yet.
β Does not pass accuracy test or does not run.