ModelScope logo How to Get a Free ModelScope API Key (2026)

54 free models available — no credit card required. Get your ModelScope API key → Test free models →

ModelScope FreeLLM Score

✅ 69/100 Solid Choice — Strong in frontier model quality How we score →
🎁 Generosity 60 🌍 Access 65 📚 Breadth 80 ⚡ Reliability 70 🔌 Compat 50 🧠 Quality 90

All Free ModelScope Models — Context Windows & Rate Limits

Model Context Max Output Modality Rate Limit Released Status
Qwen/Qwen3.5-27B 131K 131K textimagevideoaudioreasoning 2,000 RPD total; <=500 RPD/model (dynamic) Feb 24, 2026 Online
deepseek-ai/DeepSeek-V4-Pro 8K 4K Apr 24, 2026 Online
Qwen/Qwen3.5-35B-A3B 131K 131K textimagevideoaudioreasoning 2,000 RPD total; <=500 RPD/model (dynamic) Feb 24, 2026 Online
deepseek-ai/DeepSeek-V4-Flash 8K 4K Apr 24, 2026 Online
Qwen/Qwen3.5-27B 8K 4K Feb 25, 2026 Online
Qwen/Qwen3.5-35B-A3B 8K 4K Feb 25, 2026 Online
deepseek-ai/DeepSeek-V3.2 8K 4K Dec 1, 2025 Online
Qwen/Qwen3.5-122B-A10B 8K 4K Feb 25, 2026 Online
Qwen/Qwen3.5-397B-A17B 8K 4K Feb 16, 2026 Online
LLM-Research/c4ai-command-r-plus-08-2024 8K 4K Online
Qwen/Qwen3-235B-A22B-Thinking-2507 8K 4K reasoning Jul 25, 2025 Online
GLM-4.7-FlashX 200K 131K reasoning Online
Qwen/Qwen3-Next-80B-A3B-Thinking 8K 4K reasoning Sep 11, 2025 Online
GLM-5.2 1.0M 131K reasoning Online
MiniMax-M3 512K 128K visionreasoning Online
opencompass/CompassJudger-1-32B-Instruct 8K 4K Online
stepfun-ai/Step-3.5-Flash 8K 4K Online
stepfun-ai/Step-3.7-Flash 8K 4K Online
XGenerationLab/XiYanSQL-QwenCoder-32B-2412 8K 4K Online
XGenerationLab/XiYanSQL-QwenCoder-32B-2504 8K 4K Online
Kimi K2.5 262K 262K visionreasoning Online
MiniMax-M2.5-highspeed 205K 131K reasoning Online
GLM-5.1 200K 131K reasoning Online
GLM-5.1 200K 131K reasoning Online
Qwen/Qwen3-235B-A22B-Instruct-2507 8K 4K Jul 21, 2025 Online
Qwen/Qwen3-VL-235B-A22B-Instruct 8K 4K Sep 23, 2025 Online
Qwen/Qwen3-Coder-30B-A3B-Instruct 8K 4K Jul 31, 2025 Online
LLM-Research/Llama-4-Maverick-17B-128E-Instruct 8K 4K Online
MedAIBase/AntAngelMed 8K 4K Online
MiniMax/MiniMax-M1-80k 8K 4K Online
mistralai/Mistral-Small-Instruct-2409 8K 4K Online
MusePublic/Qwen-Image-Edit 8K 4K Online
OpenGVLab/InternVL3_5-241B-A28B 8K 4K Online
PaddlePaddle/ERNIE-4.5-21B-A3B-PT 8K 4K Online
PaddlePaddle/ERNIE-4.5-300B-A47B-PT 8K 4K Online
PaddlePaddle/ERNIE-4.5-VL-28B-A3B-PT 8K 4K Online
Qwen/Qwen-Image-Edit 8K 4K Online
Qwen/Qwen3-4B 8K 4K Online
Shanghai_AI_Laboratory/Intern-S1 8K 4K Online
Shanghai_AI_Laboratory/Intern-S2-Preview 8K 4K Online
Qwen/Qwen3-30B-A3B-Thinking-2507 8K 4K reasoning Aug 28, 2025 Online
Qwen/Qwen3-Next-80B-A3B-Instruct 8K 4K Sep 11, 2025 Online
meituan-longcat/LongCat-Flash-Lite 8K 4K Online
mistralai/Ministral-8B-Instruct-2410 8K 4K Online
Shanghai_AI_Laboratory/Intern-S1-mini 8K 4K Online
Qwen/Qwen3-VL-8B-Thinking 8K 4K reasoning Oct 14, 2025 Online
Qwen/Qwen3-235B-A22B 8K 4K Apr 28, 2025 Online
Qwen/Qwen3-VL-8B-Instruct 8K 4K Oct 14, 2025 Online
mistralai/Mistral-Large-Instruct-2407 8K 4K Nov 19, 2024 Online
Qwen/Qwen3-30B-A3B 8K 4K Apr 28, 2025 Online
Qwen/Qwen3-8B 8K 4K Apr 28, 2025 Online
PaddlePaddle/ERNIE-4.5-0.3B-PT 8K 4K Online
Qwen/Qwen3-14B 8K 4K Apr 28, 2025 Online
Qwen/Qwen3-32B 8K 4K Apr 28, 2025 Online

What is ModelScope?

Alibaba ModelScope — Qwen3.5, DeepSeek at 2,000 RPD free.

ModelScope is Alibaba's model hub and inference platform, providing free API access to Qwen, DeepSeek, and other models popular in the Chinese AI ecosystem. The free tier allows 2,000 requests/day total with a per-model cap of 500 RPD. Best latency for China-based users.

  • 2,000 RPD total, 500 RPD per model
  • Qwen3.5, DeepSeek available
  • Best latency in China
  • OpenAI-compatible endpoint

API Compatibility: OpenAI SDK-compatible (Chat Completions)

How to Get a ModelScope API Key

  1. 1
    Sign up at modelscope.cn Alibaba account or phone. No credit card.
  2. 2
    Go to Access Tokens
  3. 3
    Create an access token
  4. 4
    Choose a model Qwen3.5 and DeepSeek available. 2,000 RPD total.
  5. 5
    Configure OpenAI client Base URL: https://api-inference.modelscope.cn/v1

ModelScope Free Tier Limits & Pricing

Credit Card Not required
Free Tier Permanently free
Context Range 8K – 1.0M
Total Models 54 free
Rate Limits 2,000 RPD total; <=500 RPD/model (dynamic)
API Compatibility OpenAI SDK-compatible (Chat Completions)

ModelScope API Setup Tutorial & Tools

ModelScope is fully compatible with popular AI coding assistants like Cursor, Claude Code, and more. To see step-by-step API configuration instructions for your favorite tool, please visit our Global Configuration Guide →

Use Cases

What ModelScope's free models are best for, based on aggregated model capabilities:

Chat 54 models

Limitations & Caveats

  • China-hosted — high latency outside Asia-Pacific
  • 500 RPD per-model cap, 2,000 RPD total across all models
  • Alibaba account or Chinese phone required

Frequently Asked Questions

What's the difference between ModelScope and SiliconFlow?

ModelScope is Alibaba's official model hub with Qwen3.5 as the flagship. SiliconFlow is an independent platform. Both are China-hosted, but ModelScope has official Qwen models while SiliconFlow offers a broader mix including DeepSeek.

Does ModelScope have the latest Qwen models?

Yes — as Alibaba's own platform, ModelScope gets the latest Qwen releases (currently Qwen3.5) before they appear on other providers.

Can international users use ModelScope?

ModelScope requires an Alibaba account or Chinese phone for registration. International users may face difficulty signing up. Consider OpenRouter or Groq as alternatives for Qwen models.

See our FAQ for common questions about free LLM APIs