ZygmaBeta
All systems operational
Loading deployments…
1Select a model
text8B131k ctx
Llama 3.1 8B Instruct
Meta
Fast and efficient instruct-tuned model with strong performance across benchmarks
generallightweightmultilingual
text8.2B32k ctx
Qwen3 8B
Qwen
Dense model with thinking/non-thinking modes for reasoning and dialogue
reasoningcodingmultilingual
2Choose a speed tier
3Deploy