--- title: "ModelScope API-Inference free inference" description: "No-cost research and development inference across 1 model; not unrestricted production use." canonical_url: "https://freeinferencing.com/provider/modelscope_inference/" md_url: "https://freeinferencing.com/provider/modelscope_inference.md" last_updated: "2026-09-19" --- # ModelScope API-Inference > No-cost research and development inference across 1 model; not unrestricted production use. ## Classification - **Directory:** Current - **Free access:** Research / development - **Status:** Restricted access - **Confidence:** High - **Payment card:** Unknown - **Account:** Required - **Equivalent paid value:** At least $461.79/mo - **API endpoint:** `https://api-inference.modelscope.cn/v1/` ## Models mentioned - `Qwen/Qwen3.5-35B-A3B` ## Limits and terms ```yaml calls_per_day_per_account: 2000 calls_per_model_per_day: 200 selected_expensive_models_calls_per_day: 100 concurrency: Dynamic and model-specific. ``` ## What happens to your prompts? **Partially private.** API-specific retention and training protections are not documented clearly enough for a stronger classification. Privacy is audited separately from price. Review the current governing terms before sending sensitive or regulated data. ## Data governance and agreements ```yaml data_governance: review_status: partial plan_scope: ModelScope account and hosted API-Inference service; individual model licenses can add restrictions. prompt_retention: not_documented response_retention: not_documented ordinary_logging: The general platform privacy policy covers account, device, and usage information, but no API-content logging duration was found. model_training: No service-specific commitment excluding API prompts or outputs from training was found in the reviewed public documents. product_improvement: General platform terms permit service operation and improvement uses; API-content scope is not clearly separated. human_or_operator_access: not_documented subprocessors_and_routing: ModelScope and infrastructure providers process requests; model-specific licenses and publishers may create additional trust boundaries. deletion_controls: General privacy requests exist, but no API prompt/output deletion control or TTL was found. caveat: General Chinese-language platform policies exist, but the hosted inference documentation does not publish a clear prompt-retention or training matrix. agreements: terms_of_service: https://modelscope.cn/terms privacy_policy: https://modelscope.cn/privacy ``` ## Eligibility ```yaml account_required: true alibaba_cloud_link_required: true real_name_verification_required: true allowed_use: noncommercial_and_nonprofit ``` ## API compatibility and modalities - **Compatibility:** OpenAI Compatible - **Modalities:** Model inference ## Before you build with ModelScope API-Inference ### Read the classification narrowly This record describes the bounded offer supported by evidence on 2026-09-19; it does not guarantee permanence, production suitability, uptime, latency, model quality, or access from every account and region. “Research / development” should be interpreted together with the Current directory placement, High confidence, Unknown payment-card status, and the plan-specific sources below. ### Resolve the live model route The snapshot records 1 model ID. Match the exact ID against the provider's current catalog before using it in code because zero-price routes, aliases, context limits, and feature support can rotate while an older documentation page remains online. ### Match the quota to the workload shape Translate the allowance into peak requests per minute, input and output tokens, concurrency, retries, and every daily or monthly ceiling that applies to the intended account. The first limit reached by the workload is the practical ceiling. A large token pool can still fail interactive bursts, and a high request limit can still fail long-context work. Include tool calls and retry traffic, test the largest realistic payload, and treat research, experimental, and community access as conditional on their eligibility and fair-use terms. ### Preserve billing and privacy boundaries The recorded payment-card requirement is Unknown, and account access is required. Confirm both in the live signup flow, set provider-side budgets when charges are possible, and observe the balance or usage fields after a complete request. The prompt-handling classification is **Partially private** because API-specific retention and training protections are not documented clearly enough for a stronger classification. Re-read the terms for the exact route and plan before sending sensitive, regulated, or proprietary content. ### Follow the evidence, then re-check it This record links 12 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest and most specific governing document or live catalog when sources disagree. A dated finding can become stale even when the page remains online, so retain the source and snapshot that supported the decision and submit a correction when a provider changes a material term. ### Plan fallback without policy drift A fallback should preserve modality, context length, streaming, structured output, tools, safety controls, and data terms, not only API syntax. Decide which errors may retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. Rotating aliases and free pools can change behavior without changing the endpoint, so retain the response model field and test at least one substitute route before the primary offer becomes unavailable. ### Monitor the offer as a dependency Capture the model ID, response model, rate-limit headers, usage fields, latency, HTTP status, and provider request identifier. Watch authorization failure, quota exhaustion, catalog removal, policy revision, and balance movement as separate failure modes. Re-check the live catalog and governing sources on a schedule proportionate to the workload's importance, and keep an owner and exit path for any production dependency on volatile free capacity. ### Test one complete request before scaling Start with the smallest permitted request using the exact credential, model ID, endpoint, region, and account type intended for deployment. Record the status, headers, usage fields, response model, latency, and dashboard balance movement. Then exercise an invalid model, quota exhaustion, or rate limit so failure is explicit and cannot silently switch to a paid route. Validate streaming, structured output, and tool calls separately because a free model can expose fewer features than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and do not send sensitive data until the observed route matches the reviewed agreement. ## Primary sources - [API-Inference limits](https://modelscope.cn/docs/model-service/API-Inference/limits): Official Docs - [API-Inference introduction](https://modelscope.cn/docs/model-service/API-Inference/intro): Official Docs - [Free API-Inference resources](https://www.modelscope.cn/learn/1409): Official Product - [ModelScope Terms](https://modelscope.cn/terms): Official Terms - [ModelScope Privacy Policy](https://modelscope.cn/privacy): Official Privacy - [Qwen3.5-35B-A3B model card (context and generation length)](https://huggingface.co/Qwen/Qwen3.5-35B-A3B): Model Card - [OpenRouter qwen/qwen3.5-35b-a3b](https://openrouter.ai/qwen/qwen3.5-35b-a3b): Router Pricing - [OpenRouter deepseek/deepseek-r1-0528](https://openrouter.ai/deepseek/deepseek-r1-0528): Router Pricing - [OpenRouter deepseek/deepseek-v3.2-exp](https://openrouter.ai/deepseek/deepseek-v3.2-exp): Router Pricing - [Alibaba Cloud developer community ModelScope page citing 14M+ developers and 50K+ models](https://developer.aliyun.com/modelscope/): Official Product - [PyPI download stats for modelscope](https://pypistats.org/packages/modelscope): Package Registry Stats - [ModelScope primary repository](https://github.com/modelscope/modelscope): Official Repository ## Sitemap See the full [semantic sitemap](/sitemap.md) for every page and markdown mirror.