Free LLM API Service

Available Models:

• deepseek-r1:1.5b (20 endpoints)
• qwen2.5:1.5b (2 endpoints)
• smollm2:135m (54 endpoints)
• llama3.2:3b-instruct-q5_K_M (29 endpoints)
• llama3.2:latest (7 endpoints)
• mario:latest (4 endpoints)
• qwen2.5vl:3b (2 endpoints)
• llama3.2:3b (3 endpoints)
• qwen2.5:0.5b (3 endpoints)
• deepseek-r1:latest (2 endpoints)
• deepseek-r1:8b (4 endpoints)
• deepseek-r1:7b (6 endpoints)
• yi:9b (1 endpoints)
• DeepSeekR1_1.5B:latest (1 endpoints)
• deepseek-coder:1.3b (2 endpoints)
• qwen:0.5b (1 endpoints)
• llama3.2:1b (2 endpoints)
• qwen2.5:14b (1 endpoints)
• qwen2.5:32b (3 endpoints)
• qwen2.5:3b (2 endpoints)
• llama3.1:8b-instruct-q4_K_M (4 endpoints)
• llama3-backup:latest (4 endpoints)
• llama3.1:8b (4 endpoints)
• deepseek-r1:32b (2 endpoints)
• phi3:latest (1 endpoints)
• gemma2:latest (1 endpoints)
• qwen3:8b (2 endpoints)
• qwen3:0.6b (2 endpoints)
• gemma3:1b (2 endpoints)
• qwen2.5vl:32b (2 endpoints)
• qwen3:32b (2 endpoints)
• deepseek-r1:14b (2 endpoints)
• tinyllama:1.1b (1 endpoints)
• DeepSeek-R1-Distill-Qwen-32B-Q4_K_M:latest (1 endpoints)
• DeepSeek-R1-Distill-Qwen-7B-Q4_K_M:latest (1 endpoints)
• qwen3:1.7b (1 endpoints)
• minicpm-v:latest (1 endpoints)
• tinyllama:latest (1 endpoints)
• gemma3:12b (1 endpoints)
• qwen3:4b (1 endpoints)
• qwen2.5:7b (1 endpoints)
• llama3:latest (1 endpoints)
• mistral:latest (1 endpoints)
• llama3.2:1b-instruct-q4_K_M (1 endpoints)
• phi4-mini:latest (1 endpoints)

Get API Key

Usage

Base URL: http://zenith-llm.fr.to/v1

Use your API key in the Authorization header: Bearer YOUR_API_KEY

Rate limit: 100 requests per minute per API key

Disclaimer

Please do not abuse them. Be careful with sensitive information, as I do not own these endpoints—the upstream provider can see what you are requesting. I do not guarantee uptime, and they are provided as-is, with no warranties. I also do not guarantee the speed of these endpoints. And also the quality of these endpoints. I also dont guarantee the uptime.