Catalog
Open-source AI models, and the jurisdiction each one runs under
Frontière AI serves 27 models live out of 28 in the catalog, behind a single OpenAI-compatible API. 14 of them run on EU-sovereign infrastructure — providers with no non-EU capital control, designed to stay outside US jurisdiction. The others are labeled “fast access” because today they are only reachable through US-controlled infrastructure, and that label appears on the card and in the API response before you call. Every price below is what you actually pay, per 100,000 tokens, margin included — the entry price is €0.0056 on GPT-OSS 20B (OpenAI). No subscription: you top up a prepaid balance from €10 and calls stop at zero.
27
models live, 28 in catalog
14
run on EU sovereign infrastructure
€0.0056
entry price /100k input tokens
Best for…
Which model for which job
Eight categories, each derived from the registry with a stated rule — not editorial picks. The rule is printed under every card so you can recompute the answer yourself.
Qwen3.5 9B (fast, budget)
Fastest response — smallest live model
GPT-OSS 20B (OpenAI)
Cheapest — lowest combined input + output price
Kimi K3 (2.8T parameters, 1M context)
Best reasoning — largest live model flagged “Reasoning”
Qwen3 Coder 30B
Best coding — code-specialised model
Qwen2.5-VL 72B (vision)
Best vision — dedicated vision model
GPT-OSS 20B (OpenAI)
Best agents — measured tool calling, sovereign first
GLM-5.3 Flash (Zhipu/Z.ai)
Best long context — largest verified context window
GLM-5.2 (Zhipu/Z.ai)
Best sovereign — most capable EU-sovereign model
Which model should I use?
Pick what matters; Frontière AI narrows the catalog to three. Two answers are enough — the rules are the ones printed above.
Budget
Sovereignty requirement
Context needs
Main use
EU sovereign
14No non-EU capital or jurisdictional control over the company operating the infrastructure — the SecNumCloud principle. No US parent, no US-controlled subsidiary: a US court order has no direct line to these providers.
| Model | Publisher | Size | Licence | Hosted via | Price /100k in | Price /100k out |
|---|---|---|---|---|---|---|
| Qwen3 235B (instruct)qwen3-235bOpen weights · GPAIModel card ↗ | Qwen | 235B (22B active) | — | 🇪🇺 scaleway | €0.105 | €0.315 |
| Qwen3.5 397B (multimodal)qwen3.5-397bmultimodalOpen weights · GPAI✓ VerifiedModel card ↗ | Qwen | 403B (17B active) | Apache-2.0 | 🇪🇺 scaleway | €0.084 | €0.504 |
| GLM-5.2 (Zhipu/Z.ai)glm-5.2Open weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | Zhipu / Z.ai | 753B | MIT | 🇪🇺 scaleway | €0.252 | €0.77 |
| DeepSeek V4 Flash 0731 (1M context)deepseek-v4-flash-0731context 1MOpen weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | DeepSeek | 158B (13B active) | MIT | 🇪🇺 scaleway | €0.056 | €0.112 |
| Llama 3.3 70Bllama-3.3-70bOpen weights · GPAI5/6Model card ↗ | Meta | 70B | — | 🇪🇺 ovhcloud | €0.0898 | €0.0898 |
| Qwen3.6 27B (multimodal)qwen3.6-27bmultimodalOpen weights · GPAI✓ VerifiedModel card ↗ | Qwen | 27.8B | Apache-2.0 | 🇪🇺 ovhcloud | €0.0574 | €0.3878 |
| Qwen3.5 9B (fast, budget)qwen3.5-9bmultimodalOpen weights · GPAI✓ VerifiedModel card ↗ | Qwen | 9.65B | Apache-2.0 | 🇪🇺 ovhcloud | €0.014 | €0.0224 |
| Qwen3 32Bqwen3-32bOpen weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | Qwen | 32.8B | Apache-2.0 | 🇪🇺 ovhcloud | €0.0112 | €0.0308 |
| Qwen3 Coder 30Bqwen3-coder-30bcodeOpen weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | Qwen | 30.5B (3B active) | Apache-2.0 | 🇪🇺 ovhcloud | €0.0084 | €0.0322 |
| Qwen2.5-VL 72B (vision)qwen2.5-vl-72bvisionOpen weights · GPAI5/6Model card ↗ | Qwen | 73.4B | Qwen licence (HF "other")Not a permissive licence (neither Apache nor MIT) — flag it if you run a legal review. | 🇪🇺 ovhcloud | €0.1232 | €0.1232 |
| GPT-OSS 120B (OpenAI)gpt-oss-120bOpen weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | OpenAI | 120B | Apache-2.0 | 🇪🇺 ovhcloud | €0.0112 | €0.0574 |
| GPT-OSS 20B (OpenAI)gpt-oss-20bOpen weights · GPAI✓ VerifiedModel card ↗Training docs ↗ | OpenAI | 20B | Apache-2.0 | 🇪🇺 ovhcloud | €0.0056 | €0.0224 |
| Mistral Small 3.2 24Bmistral-small-3.2-24bOpen weights · GPAI✓ VerifiedModel card ↗ | Mistral AI (France) | 24B | Apache-2.0 | 🇪🇺 ovhcloud | €0.0126 | €0.0378 |
| Qwen3.8 27B (multimodal)qwen3.8-27bmultimodalcontext 256kOpen weights · GPAI5/6Model card ↗ | Qwen | 27.8B | Apache-2.0 | 🇪🇺 ovhcloud+1 provider (🇪🇺 souverain dispo) | €0.056 | €0.378 |
Fast access (US)
13Served through Modal, a US company — subject to the CLOUD Act even on its EU region. We list these models because they are the newest frontier weights and you may want them anyway; we never call them sovereign, and a sovereign variant is tracked separately below when one is planned.
| Model | Publisher | Size | Licence | Hosted via | Price /100k in | Price /100k out |
|---|---|---|---|---|---|---|
| Kimi K3 (2.8T parameters, 1M context)kimi-k3context 1MOpen weights · GPAI5/6Model card ↗ | Moonshot AI | 2.8T | — | 🇺🇸 modal | €0.3654 | €1.8242 |
| Muse Spark 1.1 (1M context)muse-spark-1.1multimodalcontext 1MOpen weights · GPAIModel card ↗ | Meta | — | — | 🇺🇸 openrouter | €0.1512 | €0.5138 |
| Muse Spark 1.2 (1M context)muse-spark-1.2multimodalcontext 1MOpen weights · GPAIModel card ↗ | Meta | — | — | 🇺🇸 openrouter | €0.1512 | €0.5138 |
| GLM-5.3 Flash (Zhipu/Z.ai)glm-5.3-flashmultimodalcontext 1.3MOpen weights · GPAIModel card ↗Training docs ↗ | Zhipu / Z.ai | — | MIT | 🇺🇸 modal+1 provider | €0.0098 | €0.0308 |
| Qwen3.8 Max (2.4T, 1M context)qwen3.8-maxcontext 1MOpen weights · GPAIModel card ↗ | Qwen | 2.4T (95B active) | Qwen licence (proprietary)Not a permissive licence (neither Apache nor MIT) — flag it if you run a legal review. | 🇺🇸 modal+1 provider | €0.2422 | €0.7252 |
| Qwen3.8 Flash (Qwen)qwen3.8-flashcontext 128kOpen weights · GPAIModel card ↗ | Qwen | — | Qwen licence (proprietary)Not a permissive licence (neither Apache nor MIT) — flag it if you run a legal review. | 🇺🇸 openrouter | €0.0182 | €0.0574 |
| DeepSeek R1 (671B MoE)deepseek-r1context 128kOpen weights · GPAIModel card ↗Training docs ↗ | DeepSeek | 671B (39B active) | MIT | 🇺🇸 openrouter | €0.0854 | €0.3024 |
| DeepSeek V4 Pro (345B MoE)deepseek-v4-procontext 128kOpen weights · GPAIModel card ↗ | DeepSeek | 345B (44B active) | MIT | 🇺🇸 openrouter | €0.1064 | €0.21 |
| DeepSeek R1 (Distill Qwen 32B)deepseek-r1-distill-qwen-32bcontext 128kOpen weights · GPAIModel card ↗ | DeepSeek | 32B | MIT | 🇺🇸 openrouter | €0.0154 | €0.0616 |
| DeepSeek R1 (Distill Qwen 14B)deepseek-r1-distill-qwen-14bcontext 128kOpen weights · GPAIModel card ↗ | DeepSeek | 14B | MIT | 🇺🇸 openrouter | €0.0098 | €0.0252 |
| GLM-5.3 (Zhipu/Z.ai)glm-5.3multimodalcontext 1.3MOpen weights · GPAIModel card ↗Training docs ↗ | Zhipu / Z.ai | — | MIT | 🇺🇸 openrouter | €0.1694 | €0.532 |
| Gemma 3 27B (Google)gemma-3-27bcontext 128kOpen weights · GPAIModel card ↗ | 27B | Apache-2.0 | 🇺🇸 openrouter | €0.0098 | €0.0546 | |
| DeepSeek V4.1 Flash (552B MoE, 1M context)deepseek-v4.1-flashmultimodalcontext 1MOpen weights · GPAI✓ VerifiedModel card ↗ | DeepSeek | 552B (8B prefill / 16B decode active) | MIT | 🇺🇸 modal | €0.0182 | €0.0728 |
Coming soon
1Entries already validated (weights public, provider identified) but not callable yet. They carry no price until a real one is measured — a model is never switched live with a placeholder tariff.
| Model | Publisher | Size | Licence | Hosted via |
|---|---|---|---|---|
| Kimi K3 — EU sovereign variantkimi-k3-souveraincontext 1MOpen weights · GPAIModel card ↗ | Moonshot AI | 2.8T | — | 🇪🇺 self-hosted |
A dash means we have not verified that value ourselves — not that it does not exist. We would rather leave a gap than publish a licence or a parameter count we have not checked against the publisher.
A curated selection, not every open-source model
An entry is added only once we have verified who serves it, under which licence, and at what real cost — through the provider's own API, cross-checked against the Hugging Face Hub. That is why this catalog is short and why the numbers on this page are the ones you will be billed.
How we draw the sovereignty line: GDPR vs. the CLOUD Act →FAQ
Which models on Frontière AI are GDPR-compliant?
The 14 models labeled EU sovereign. They run on OVHcloud, Scaleway or our own dedicated EU servers — companies with no non-EU capital control, so a US court order has no direct line to them. Models labeled “fast access” are served through US-controlled infrastructure and are marked as such before you call.
What is the cheapest model on Frontière AI?
GPT-OSS 20B (OpenAI), at €0.0056 per 100,000 input tokens — margin included, that is the amount actually debited from your prepaid balance. Output tokens are billed separately at each model's own rate, listed in the tables above.
Can I switch models without changing my code?
Yes. Every model is reachable through the same OpenAI-compatible endpoint — you change the “model” field in the request body, nothing else. One API key covers the whole catalog, instead of one account per model vendor.
Why is Kimi K3 not labeled EU sovereign?
Because it is served through Modal, a US company, and a US company is subject to the CLOUD Act even when it runs in an EU region — physical location is not what decides jurisdiction. An EU sovereign variant of Kimi K3, self-hosted on dedicated EU servers, is listed separately as coming soon.
Call any of these models in minutes
Create an account, top up €10, change two lines. One key for the whole catalog.