ModelScope is an open-source AI platform providing access to a wide range of models, datasets, and tools for development and deployment. It supports tasks across computer vision (e.g., visual inspection, OCR, face recognition), natural language processing (e.g., text generation, translation, summarization), speech (e.g., speech recognition, synthesis), and multimodal applications (e.g., image description, text-to-image). Key models include Z-Image-Turbo, DeepSeek-V3.2, GLM-4.6, and Mistral-Large-3-675B-Instruct-2512. The platform features ModelScope Library for Python-based inference and fine-tuning, ModelHub for hosting models and datasets, and Studios for building and showcasing AI applications. It focuses on community-driven development with trending models and datasets updated regularly.
ModelScope provides model inference API access to a wide range of open-source AI models via OpenAI-compatible endpoints.
- Models
- 82 models
- From
- --
- Speed
- 44 tok/s
- Updated
- 7/8/2026
- Latency
- 0.00 s
- Created At
- 12/7/2025
API Endpoints
- Endpoint 1
api-inference.modelscope.cn - Endpoint 2
ms-ens-1f4a9445-d0e7.api-inference.modelscope.cn - Endpoint 3
ms-ens-327b7543-e27c.api-inference.modelscope.cn - Endpoint 4
ms-ens-51815792-b22c.api-inference.modelscope.cn - Endpoint 5
ms-ens-257c1f27-2cc6.api-inference.modelscope.cn
Verify ownership to unlock provider management features:
- Edit provider name, content, and links
- Get featured with priority traffic and visibility boost
- Display a verified badge to build user trust
Leaderboard Rankings
Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.
About ModelScope
Health Check
API Benchmarks & Pricing
Compare 81 model rows across audit recency, latest speed tests, throughput, latency, and per-token pricing.
Qwen/Qwen3-235B-A22B-Instruct-2507 | - | - |
Qwen/Qwen3-4B | - | - |
Qwen/Qwen3-Coder-30B-A3B-Instruct | - | - |
Qwen/Qwen3-Next-80B-A3B-Instruct | - | - |
Qwen/Qwen3-Embedding-0.6B | - | - |
Qwen/Qwen3-Embedding-4B | - | - |
Qwen/Qwen3-Embedding-8B | - | - |
Qwen/Qwen3.8-Flash-Next | - | - |
deepseek-ai/DeepSeek-V4-Pro-0813 | - | - |
early-access/EA-29B-A4B | - | - |
MiniMax/MiniMax-M3 | - | - |
Qwen/Qwen3.8-27B | 85.4 t/s | 2.52 s |
stepfun-ai/Step-3.7-Flash | - | - |
Tencent-Hunyuan/Hy3 | - | - |
| 138.5 t/s | 12.64 s | |
deepseek-ai/DeepSeek-R1-0528 | - | - |
deepseek-ai/DeepSeek-V4-Pro | - | - |
MedAIBase/AntAngelMed | - | - |
Shanghai_AI_Laboratory/Intern-S2-Preview | - | - |
deepseek-ai/DeepSeek-R1-Distill-Llama-8B | - | - |
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B | - | - |
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B | - | - |
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B | - | - |
deepseek-ai/DeepSeek-V4-Flash | - | - |
iic/GUI-Owl-1.5-8B-Instruct | - | - |
Showing 25 of 81 model rows
Recent Test Records
| Time | Model | Speed | Latency |
|---|---|---|---|
| Sep 2, 04:13 AM | ZhipuAI/GLM-5.2 | 138.53 tok/s | 12.64s |
| Sep 2, 04:12 AM | deepseek-ai/DeepSeek-V4-Flash-0731 | 92.81 tok/s | 3.03s |
| Aug 24, 07:53 PM | Qwen/Qwen3.8-27B | 85.40 tok/s | 2.52s |
| Mar 30, 04:09 AM | kgiser/gpu_gpt_5 | 26.05 tok/s | 1.11s |
| Mar 30, 03:24 AM | kgiser/gpu_gpt_5 | 30.59 tok/s | 1.17s |
| Mar 30, 03:16 AM | TeichAI/Qwen3-30B-A3B-Thinking-2507-Claude-4.5-Sonnet-High-Reasoning-Distill-GGUF | 13.07 tok/s | 2.17s |
| Mar 30, 02:41 AM | Qwen/QVQ-72B-Preview | 44.75 tok/s | 1.16s |
| Mar 30, 02:39 AM | Qwen/Qwen3-Coder-480B-A35B-Instruct | 61.17 tok/s | 0.94s |
| Mar 30, 02:37 AM | moonshotai/Kimi-K2.5 | 43.12 tok/s | 1.11s |
| Mar 30, 02:32 AM | MiniMax/MiniMax-M2.5 | 52.75 tok/s | 7.49s |
Similar API Provider Alternatives to Compare
Compare ModelScope alternatives against 6 nearby API providers using 228 LMSpeed signals across shared model coverage, pricing, benchmark speed, uptime, and free-model availability.
| Provider | Why compare | Models | Free | Avg price | Speed | 30d uptime |
|---|---|---|---|---|---|---|
| ModelScope api-inference-modelscope-cn ModelScope provides model inference API access to a wide range of open-source AI models via OpenAI-compatible endpoints. | Current provider baseline | 82 | 0 | N/A | 44 tok/s | 5700% |
| OpenRouter openrouter A unified API interface providing access to over 300 models from 60+ providers, including OpenAI, Anthropic, and Google. |
| 561 | 0 | N/A | 85 tok/s | 9990% |
| NVIDIA NIM nvidia-nim NVIDIA NIM provides optimized AI model inference APIs for LLMs, vision, and embedding models through NVIDIA cloud infrastructure. |
| 166 | 0 | N/A | 78 tok/s | 9960% |
| Novita AI novita-ai Provides a unified API for accessing 200+ AI models, custom model deployment, GPU cloud services, and secure agent sandboxes. |
| 142 | 0 | N/A | 56 tok/s | 6910% |
| 七牛云 qiniu-2 七牛云提供AI大模型推理服务，包括多种模型调用、智能问答助手、代码助手等企业级AI解决方案。 |
| 89 | 0 | N/A | 90 tok/s | 9980% |
| SiliconFlow siliconflow Provides cost-effective generative AI cloud services based on open-source models for text, image, video, and audio generation. |
| 154 | 0 | N/A | 59 tok/s | 4150% |
| 并行科技 llmapi-paratera-com Provides high-performance computing (HPC) and AI infrastructure services, including GPU cloud servers, MaaS platforms, and industry-specific solutions. |
| 97 | 0 | N/A | 195 tok/s | 0% |
Notes
- Health checks: Scope: the 72-hour chart and recent availability measure API connectivity only. Each bar summarizes one hour of checks. Targets: LMSpeed tries the configured health check URL and provider status URL first, then API endpoints derived from known API hosts and recent speed-test base URLs. A website host is considered only when it looks like an API endpoint. Probe steps: each candidate goes through DNS lookup, TCP connection, TLS handshake for HTTPS, and an HTTP HEAD request with redirects followed. Probing stops after the first reachable candidate. Reachable criteria: every required network step must succeed. An HTTP response below 500 is treated as reachable, including 401 because it confirms that an authenticated API endpoint responded, except for statuses classified as blocked. Blocked results: HTTP 403, 429, 521, 525, and 530, plus detected WAF or Cloudflare challenges, are shown as blocked and excluded from availability calculations because LMSpeed cannot determine whether the API itself is down. Model availability: when a dedicated test key is configured, LMSpeed sends an authenticated GET request to a derived /models endpoint and compares returned model IDs with this provider's listed models. These per-model results appear in Models & Pricing and are not included in the provider connectivity percentage. Timeouts: TCP connection, TLS handshake, HTTP connectivity, and model requests each use a 20-second timeout. A full run can take longer when several candidates are tried. Frequency: a background worker checks all providers every 5 minutes by default. The 72-hour chart combines those samples into hourly bars, and the schedule may be changed by the service operator. Limit: automated samples are not an SLA and do not guarantee account quota, every model, every region, or successful completion requests. Check the provider's own status page before making operational decisions.
- Domain Rating data is sourced from Ahrefs. It is a 0–100 backlink-based domain strength signal and does not measure API speed or reliability.
- Announcements and FAQ are read from this provider's NewAPI status snapshot when available. LMSpeed stores the original content and optional English translations from the provider status source, then shows the localized fields on this page.