ModelScope

Inference layers 53 models data_array 1M max context Free tier

info Notes

Free API-Inference for registered users. Requires Alibaba Cloud account binding + real-name verification. []

key How to get API key

General steps — follow the official ModelScope page for exact details.

Open the ModelScope API key page ↗
Create a new account, or sign in if you already have one.
Generate an API key from the dashboard or settings.

layers Available models 53

Model	Capabilities	Context	Output	Rate limit
`Qwen/Qwen3.5-35B-A3B` `qwen/qwen3.5-35b-a3b`	description	256K	—	2,000 RPD total; <=500 RPD/model (dynamic)
`Qwen/Qwen3.5-27B` `qwen/qwen3.5-27b`	description	256K	—	2,000 RPD total; <=500 RPD/model (dynamic)
+ API-Inference-enabled models `+-api-inference-enabled-models`	LLM, MLLM	Varies	—	Dynamic quotas + dynamic concurrency
Qwen 3.8 27B `qwen-3.8-27b`	description	1M	—	2,000 requests per day
DeepSeek V4 Flash `deepseek-v4-flash`	description	1M	—	2,000 requests per day
DeepSeek V4 Pro `deepseek-v4-pro`	description	1M	—	2,000 requests per day
DeepSeek V4 Pro 0813 `deepseek-v4-pro-0813`	description	1M	—	2,000 requests per day
EA 29B A4B `ea-29b-a4b`	description	1M	—	2,000 requests per day
AntAngelMed `antangelmed`	description	1M	—	2,000 requests per day
LongCat Flash Lite `longcat-flash-lite`	description	1M	—	2,000 requests per day
MiniMax M1 `minimax-m1`	description	1M	—	2,000 requests per day
MiniMax M3 `minimax-m3`	description	1M	—	2,000 requests per day
Mistral Large `mistral-large`	description	1M	—	2,000 requests per day
Qwen Image Edit `qwen-image-edit`	description	1M	—	2,000 requests per day
CompassJudger 32B `compassjudger-32b`	description	1M	—	2,000 requests per day
InternVL 3.5 `internvl-3.5`	description	1M	—	2,000 requests per day
ERNIE 4.5 0.3B `ernie-4.5-0.3b`	description	1M	—	2,000 requests per day
ERNIE 4.5 21B `ernie-4.5-21b`	description	1M	—	2,000 requests per day
ERNIE 4.5 300B `ernie-4.5-300b`	description	1M	—	2,000 requests per day
ERNIE 4.5 VL 28B `ernie-4.5-vl-28b`	description	1M	—	2,000 requests per day
Qwen Image Edit `qwen-image-edit`	description	1M	—	2,000 requests per day
Qwen 3 14B `qwen-3-14b`	description	1M	—	2,000 requests per day
Qwen 3 235B `qwen-3-235b`	description	1M	—	2,000 requests per day
Qwen 3 235B `qwen-3-235b`	description	1M	—	2,000 requests per day
Qwen 3 235B Thinking `qwen-3-235b-thinking`	description	1M	—	2,000 requests per day
Qwen 3 30B A3B `qwen-3-30b-a3b`	description	1M	—	2,000 requests per day
Qwen 3 30B A3B Thinking `qwen-3-30b-a3b-thinking`	description	1M	—	2,000 requests per day
Qwen 3 4B `qwen-3-4b`	description	1M	—	2,000 requests per day
Qwen 3 8B `qwen-3-8b`	description	1M	—	2,000 requests per day
Qwen 3 Coder 30B `qwen-3-coder-30b`	description	1M	—	2,000 requests per day
Qwen 3 Embedding 0.6B `qwen-3-embedding-0.6b`	description	1M	—	2,000 requests per day
Qwen 3 Embedding 4B `qwen-3-embedding-4b`	description	1M	—	2,000 requests per day
Qwen 3 Embedding 8B `qwen-3-embedding-8b`	description	1M	—	2,000 requests per day
Qwen 3 Next 80B `qwen-3-next-80b`	description	1M	—	2,000 requests per day
Qwen 3 Next 80B Thinking `qwen-3-next-80b-thinking`	description	1M	—	2,000 requests per day
Qwen 3 VL 235B `qwen-3-vl-235b`	description	1M	—	2,000 requests per day
Qwen 3 VL 8B `qwen-3-vl-8b`	description	1M	—	2,000 requests per day
Qwen 3 VL 8B Thinking `qwen-3-vl-8b-thinking`	description	1M	—	2,000 requests per day
Qwen 3.5 122B `qwen-3.5-122b`	description	1M	—	2,000 requests per day
Qwen 3.5 27B `qwen-3.5-27b`	description	1M	—	2,000 requests per day
Qwen 3.5 35B `qwen-3.5-35b`	description	1M	—	2,000 requests per day
Qwen 3.5 397B `qwen-3.5-397b`	description	1M	—	2,000 requests per day
Qwen 3.8 Flash Next `qwen-3.8-flash-next`	description	1M	—	2,000 requests per day
Intern S1 `intern-s1`	description	1M	—	2,000 requests per day
Intern S1 Mini `intern-s1-mini`	description	1M	—	2,000 requests per day
Intern S2 `intern-s2`	description	1M	—	2,000 requests per day
Step 3.5 Flash `step-3.5-flash`	description	1M	—	2,000 requests per day
Step 3.7 Flash `step-3.7-flash`	description	1M	—	2,000 requests per day
Hunyuan Hy3 `hunyuan-hy3`	description	1M	—	2,000 requests per day
XiYanSQL 32B `xiyansql-32b`	description	1M	—	2,000 requests per day
XiYanSQL 32B 2504 `xiyansql-32b-2504`	description	1M	—	2,000 requests per day
GLM 4.7 Flash `glm-4.7-flash`	description	1M	—	2,000 requests per day
GLM 5.2 `glm-5.2`	description	1M	—	2,000 requests per day

help Frequently Asked Questions

How many models does ModelScope offer for free?

ModelScope offers 53 free models.

What is the largest context window available?

The largest context window is 1,000,000 tokens.

What are the rate limits?

The free tier is limited to 2,000 requests per day.