Modèles et API gratuits sur Cloudflare en 2026
Cloudflare Workers AI exécute des modèles sur son réseau edge mondial et inclut une allocation quotidienne gratuite.
30 modèlesjusqu'à 10.5M contextemeilleur : Qwen 3.8 27BMis à jour:
Cloudflare Workers AI
En ligne lors de la dernière vérification
10 000 Neurons par jour sont gratuits sur les offres Workers Free et Paid.
- Base URL
https://api.cloudflare.com/client/v4/accounts/ACCOUNT_ID/ai/v1- Compatibilité
- OpenAI Chat Completions
- Carte
- À vérifier auprès du fournisseur
- Limites
- Workers Free: 10,000 Neurons/day shared across eligible models
| # | Modèle ↕ | Indice AA ↕ | Contexte ↕ | Taille ↕ | Limite | |
|---|---|---|---|---|---|---|
| 1 | 
Qwen 3.8 27B
@cf/qwen/qwen3.8-27b |
33.7 | 262K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 2 | 
Gemma 4 26B A4B
@cf/google/gemma-4-26b-a4b-it |
16.7 | 256K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 3 | 
GLM 4.7 Flash
@cf/zai-org/glm-4.7-flash |
14.9 | 131K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 4 | 
Nemotron 3 120B
@cf/nvidia/nemotron-3-120b-a12b |
— | 256K | Très grand | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 5 | 
gpt-oss-120b
@cf/openai/gpt-oss-120b |
11.6 | 128K | Très grand | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 6 |
gpt-oss-20b
@cf/openai/gpt-oss-20b |
10.0 | 128K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 7 | 
Llama 3.3 70B Instruct Fp8 Fast
@cf/meta/llama-3.3-70b-instruct-fp8-fast |
— | 24K | Très grand | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 8 |
Llama 4 Scout
@cf/meta/llama-4-scout-17b-16e-instruct |
8.1 | 131K | Très grand | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 9 |
Qwen3 30B
@cf/qwen/qwen3-30b-a3b-fp8 |
9.8 | 33K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 10 |
QwQ 32B
@cf/qwen/qwq-32b |
9.5 | 24K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 11 | 
DeepSeek R1 Distill Qwen 32B
@cf/deepseek-ai/deepseek-r1-distill-qwen-32b |
8.4 | 80K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 12 |
Llama 3.1 8B
@cf/meta/llama-3.1-8b-instruct-fp8 |
6.9 | 32K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 13 |
Gemma SEA-LION V4 27B
@cf/aisingapore/gemma-sea-lion-v4-27b-it |
— | 128K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 14 | 
Mistral Small 3.1 24B
@cf/mistralai/mistral-small-3.1-24b-instruct |
— | 128K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 15 |
Llama 3.2 11B Vision
@cf/meta/llama-3.2-11b-vision-instruct |
5.4 | 128K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 16 |
Qwen2.5 Coder 32B
@cf/qwen/qwen2.5-coder-32b-instruct |
6.7 | 33K | Taille moyenne | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 17 |
Llama 3.2 1B
@cf/meta/llama-3.2-1b-instruct |
4.8 | 60K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 18 |
apertus-v1.5-8b
@cf/swiss-ai/apertus-v1.5-8b |
— | 262K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 19 |
phonellm-mini-alpha-2
@cf/daily-co/phonellm-mini-alpha-2 |
— | 262K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 20 |
Llama 3.2 3B
@cf/meta/llama-3.2-3b-instruct |
5.7 | 80K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 21 |
eurollm-9b-it
@cf/utter-project/eurollm-9b-it |
— | 32K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 22 |
clef
@cf/cloudflare/clef |
— | 66K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 23 |
Granite 4.0 H Micro
@cf/ibm-granite/granite-4.0-h-micro |
— | 131K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 24 |
clairic
@cf/cloudflare/clairic |
— | 66K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 25 |
clef-omni
@cf/cloudflare/clef-omni |
— | 66K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 26 |
Gemma 7B
@cf/google/gemma-7b-it-lora |
— | 4K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 27 |
clef-flash
@cf/cloudflare/clef-flash |
— | 25K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 28 |
Mistral 7B
@cf/mistral/mistral-7b-instruct-v0.2-lora |
5.0 | 15K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 29 |
Gemma 2B
@cf/google/gemma-2b-it-lora |
— | 8K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
| 30 |
Llama 2 7B
@cf/meta-llama/llama-2-7b-chat-hf-lora |
— | 8K | Petit | Workers Free: 10,000 Neurons/day shared across eligible models | Ouvrir |
Aucun modèle ne correspond à vos filtres.
Benchmarks : Artificial Analysis (24 septembre 2026). Un tiret signifie que le modèle ne figure pas dans leur classement ; nous n'estimons jamais les scores nous-mêmes.
FAQ
Cloudflare Workers AI est-il gratuit ?
10 000 Neurons par jour sont gratuits sur les offres Workers Free et Paid. 30 modèles sont actuellement gratuits.
Cloudflare Workers AI est-il compatible avec l'API OpenAI ?
Oui — utilisez la base URL https://api.cloudflare.com/client/v4/accounts/ACCOUNT_ID/ai/v1 avec n'importe quel client ou SDK compatible OpenAI.
Cloudflare Workers AI exige-t-il une carte ?
Cela dépend du compte — consultez le tableau de bord du fournisseur.