Models

The following table lists models supported by MindSpore Transformers.

Model	Specifications	Model Type	Model Architecture	Latest Version
TeleChat3 🔥HOT	36B	Dense LLM	Mcore	1.9.0
TeleChat3-MoE 🔥HOT	105B-A4.7B	Sparse LLM	Mcore	1.9.0
Qwen3 🔥HOT	0.6B/1.7B/4B/8B/14B/32B	Dense LLM	Mcore	1.9.0
Qwen3-MoE 🔥HOT	30B-A3B/235B-A22B	Sparse LLM	Mcore	1.9.0
DeepSeek-V3 🔥HOT	671B	Sparse LLM	Mcore/Legacy	1.9.0
GLM4.5 🔥HOT	106B-A12B/355B-A32B	Sparse LLM	Mcore	1.9.0
GLM4 🔥HOT	9B	Dense LLM	Mcore/Legacy	1.9.0
Qwen2.5 🔥HOT	0.5B/1.5B/7B/14B/32B/72B	Dense LLM	Legacy	1.9.0
TeleChat2 🔥HOT	7B/35B/115B	Dense LLM	Mcore/Legacy	1.9.0
Llama3.1 ⚠️EOL	8B/70B	Dense LLM	Legacy	1.7.0
Mixtral ⚠️EOL	8x7B	Sparse LLM	Legacy	1.7.0
CodeLlama ⚠️EOL	34B	Dense LLM	Legacy	1.5.0
CogVLM2-Image ⚠️EOL	19B	MM	Legacy	1.5.0
CogVLM2-Video ⚠️EOL	13B	MM	Legacy	1.5.0
DeepSeek-V2 ⚠️EOL	236B	Sparse LLM	Legacy	1.5.0
DeepSeek-Coder-V1.5 ⚠️EOL	7B	Dense LLM	Legacy	1.5.0
DeepSeek-Coder ⚠️EOL	33B	Dense LLM	Legacy	1.5.0
GLM3-32K ⚠️EOL	6B	Dense LLM	Legacy	1.5.0
GLM3 ⚠️EOL	6B	Dense LLM	Legacy	1.5.0
InternLM2 ⚠️EOL	7B/20B	Dense LLM	Legacy	1.5.0
Llama3.2 ⚠️EOL	3B	Dense LLM	Legacy	1.5.0
Llama3.2-Vision ⚠️EOL	11B	MM	Legacy	1.5.0
Llama3 ⚠️EOL	8B/70B	Dense LLM	Legacy	1.5.0
Qwen2 ⚠️EOL	0.5B/1.5B/7B/57B/57B-A14B/72B	Dense /Sparse LLM	Legacy	1.5.0
Qwen1.5 ⚠️EOL	7B/14B/72B	Dense LLM	Legacy	1.5.0
Qwen-VL ⚠️EOL	9.6B	MM	Legacy	1.5.0
TeleChat ⚠️EOL	7B/12B/52B	Dense LLM	Legacy	1.5.0
Whisper ⚠️EOL	1.5B	MM	Legacy	1.5.0
Yi ⚠️EOL	6B/34B	Dense LLM	Legacy	1.5.0
YiZhao ⚠️EOL	12B	Dense LLM	Legacy	1.5.0
Llama2 ⚠️EOL	7B/13B/70B	Dense LLM	Legacy	1.3.2
Baichuan2 ⚠️EOL	7B/13B	Dense LLM	Legacy	1.3.2
GLM2 ⚠️EOL	6B	Dense LLM	Legacy	1.3.2
GPT2 ⚠️EOL	124M/13B	Dense LLM	Legacy	1.3.2
InternLM ⚠️EOL	7B/20B	Dense LLM	Legacy	1.3.2
Qwen ⚠️EOL	7B/14B	Dense LLM	Legacy	1.3.2
CodeGeex2 ⚠️EOL	6B	Dense LLM	Legacy	1.1.0
WizardCoder ⚠️EOL	15B	Dense LLM	Legacy	1.1.0
Baichuan ⚠️EOL	7B/13B	Dense LLM	Legacy	1.0
Blip2 ⚠️EOL	8.1B	MM	Legacy	1.0
Bloom ⚠️EOL	560M/7.1B/65B/176B	Dense LLM	Legacy	1.0
Clip ⚠️EOL	149M/428M	MM	Legacy	1.0
CodeGeex ⚠️EOL	13B	Dense LLM	Legacy	1.0
GLM ⚠️EOL	6B	Dense LLM	Legacy	1.0
iFlytekSpark ⚠️EOL	13B	Dense LLM	Legacy	1.0
Llama ⚠️EOL	7B/13B	Dense LLM	Legacy	1.0
MAE ⚠️EOL	86M	MM	Legacy	1.0
Mengzi3 ⚠️EOL	13B	Dense LLM	Legacy	1.0
PanguAlpha ⚠️EOL	2.6B/13B	Dense LLM	Legacy	1.0
SAM ⚠️EOL	91M/308M/636M	MM	Legacy	1.0
Skywork ⚠️EOL	13B	Dense LLM	Legacy	1.0
Swin ⚠️EOL	88M	MM	Legacy	1.0
T5 ⚠️EOL	14M/60M	Dense LLM	Legacy	1.0
VisualGLM ⚠️EOL	6B	MM	Legacy	1.0
Ziya ⚠️EOL	13B	Dense LLM	Legacy	1.0
Bert ⚠️EOL	4M/110M	Dense LLM	Legacy	0.8

* ⚠️EOL indicates that the model has been offline from the main branch and can be used with the latest supported version (e.g., 1.7.0).