--- title: "Swiss AI Research Platform Model API free inference" description: "A recurring no-cost API allowance covering 33 cataloged models." canonical_url: "https://freeinferencing.com/provider/swiss_ai_research_platform/" md_url: "https://freeinferencing.com/provider/swiss_ai_research_platform.md" last_updated: "2026-09-19" --- # Swiss AI Research Platform Model API > A recurring no-cost API allowance covering 33 cataloged models. ## Classification - **Directory:** Current - **Free access:** Always-free quota - **Status:** Restricted access - **Confidence:** High - **Payment card:** No - **Account:** Required - **Equivalent paid value:** Not quantifiable - **API endpoint:** `https://api.swissai.svc.cscs.ch/v1` ## Models mentioned - `SwissAI-Research/Qwen/Qwen3.5-27B` - `SwissAI-Research/zai-org/GLM-4.7-Flash` - `CSCS-Inference/google/gemma-4-31B-it` - `CSCS-Inference/moonshotai/Kimi-K2.7-Code` - `CSCS-Inference/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16` - `CSCS-Inference/swiss-ai/Apertus-70B-Instruct-2509` - `CSCS-Inference/swiss-ai/Apertus-8B-Instruct-2509` - `CSCS-Inference/swiss-ai/Apertus-v1.5-70B` - `CSCS-Inference/swiss-ai/Apertus-v1.5-70B-thinking` - `CSCS-Inference/swiss-ai/Apertus-v1.5-8B` - `CSCS-Inference/swiss-ai/Apertus-v1.5-8B-thinking` - `CSCS-Inference/zai-org/GLM-5.2` - `RCP-AIaaS/BAAI/bge-reranker-v2-m3` - `RCP-AIaaS/EPFLiGHT/Apertus-70B-MeditronFO` - `RCP-AIaaS/OpenMeditron/Meditron3-70B` - `RCP-AIaaS/OpenMeditron/meditron-CHUV-dpo` - `RCP-AIaaS/Qwen/Qwen3-Embedding-8B` - `RCP-AIaaS/Qwen/Qwen3.5-397B-A17B` - `RCP-AIaaS/Qwen/Qwen3.6-35B-A3B` - `RCP-AIaaS/RedHatAI/GLM-5.3-Flash-NVFP4-nvfp4` - `RCP-AIaaS/deepseek-ai/DeepSeek-V4-Flash-0731` - `RCP-AIaaS/google/medgemma-27b-text-it` - `RCP-AIaaS/mistralai/Mistral-Small-3.2-24B-Instruct-2506` - `RCP-AIaaS/moonshotai/Kimi-K2.7-Code` - `RCP-AIaaS/openai/gpt-oss-120b` - `RCP-AIaaS/swiss-ai/Apertus-v1.5-70B` - `RCP-AIaaS/swiss-ai/Apertus-v1.5-8B` - `RCP-AIaaS/zai-org/GLM-5.3-Flash` - `dmelikidze/meta-llama/Llama-3.3-70B-Instruct` - `jiaxuzhao/Qwen/Qwen2.5-72B-Instruct` - `jiaxuzhao/deepseek-ai/DeepSeek-R1-Distill-Llama-70B` - `jiaxuzhao/meta-llama/Llama-3.1-8B-Instruct` - `omlins/deepseek-ai/DeepSeek-V4-Flash` ## Limits and terms ```yaml account_usage_budget: provisioned_at_first_sign_in_amount_not_public provider_routes_requests_per_minute: generally_less_than_15 direct_deployment_requests_per_minute: slightly_higher_numeric_value_not_public availability: shared_best_effort ``` ## What happens to your prompts? **Private.** The platform says API prompts and outputs are not stored or used for training; usage metadata may be retained indefinitely and sensitive data is prohibited. 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: reviewed plan_scope: Swiss AI Research Platform OpenAI-compatible API for academic and noncommercial research. prompt_retention: The platform FAQ says API prompts are not stored. response_retention: The platform FAQ says API outputs are not stored. ordinary_logging: Usage metadata including user identifier, timestamp, model, and token counts may be retained indefinitely. model_training: Prompts and outputs are not used for model training. product_improvement: No content-based improvement use is permitted by the stated no-storage handling; metadata can support operations and capacity planning. human_or_operator_access: Ordinary content is not retained for later review, though CSCS necessarily processes requests transiently and administers retained metadata. subprocessors_and_routing: CSCS operates the platform and its provider/direct deployments; temporary user deployments can add model-owner trust considerations within the shared service. deletion_controls: Content is not stored. No user-facing deletion control or retention maximum is published for usage metadata. caveat: The FAQ prohibits sensitive data and is an operational statement rather than a complete negotiated service agreement; indefinite metadata retention remains. agreements: operational_faq: https://serving.swissai.svc.cscs.ch/faq ``` ## Eligibility ```yaml account_required: true swiss_university_researchers: direct_institutional_sign_in external_academics: manual_access_focused_on_Apertus_related_research commercial_use_allowed: false keys_per_user: 1 payment_method_required: false ``` ## API compatibility and modalities - **Compatibility:** OpenAI Compatible - **Modalities:** Embeddings, Reranking - **Geography:** Switzerland ## Before you build with Swiss AI Research Platform Model API ### 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. “Always-free quota” should be interpreted together with the Current directory placement, High confidence, No payment-card status, and the plan-specific sources below. ### Resolve the live model route The snapshot records 33 model IDs. 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 No, 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 **Private** because The platform says API prompts and outputs are not stored or used for training; usage metadata may be retained indefinitely and sensitive data is prohibited. 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 3 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 - [Swiss AI Research Platform](https://serving.swissai.svc.cscs.ch/): Official Product - [Platform FAQ](https://serving.swissai.svc.cscs.ch/faq): Official Faq - [Public model catalog](https://api.swissai.svc.cscs.ch/v1/models): Live Catalog ## Sitemap See the full [semantic sitemap](/sitemap.md) for every page and markdown mirror.