2 ms·
Currently for chat we support Claude 2, Claude 3 Haiku, Sonnet, and Opus, GPT-4 Turbo, and Mixtral 8x7b. For autocomplete we're using Starcoder 16b. We also s
by ado__dev 2y ago
Currently for chat we support Claude 2, Claude 3 Haiku, Sonnet, and Opus, GPT-4 Turbo, and Mixtral 8x7b.
For autocomplete we're using Starcoder 16b.
We also support local inference with any LLM via Ollama (this is experimental and only for Cody Free/Pro users). Enterprise customers can choose which LLMs they want.