Skip to main content
Deprecation Notice: These models are for the deprecated Completions V1 and Chat APIs.New projects should start with Completions v2 immediately. Existing integrations should migrate as soon as possible.View Migration Guide →
The following models are available for the standard Gloo AI API endpoints (Completions V1 and Chat APIs).

​
Anthropic

Model IDMax TokensTool UseStreaming
anthropic.claude-3-5-sonnet-20240620-v1:0200,000YesYes *
us.anthropic.claude-3-5-sonnet-20240620-v1:0200,000YesYes *
us.anthropic.claude-3-5-sonnet-20241022-v2:0200,000YesYes *
us.anthropic.claude-3-opus-20240229-v1:0200,000YesYes*
anthropic.claude-3-sonnet-20240229-v1:0200,000YesYes*
anthropic.claude-3-haiku-20240307-v1:0200,000YesYes*
anthropic.claude-v2:1100,000YesYes*
anthropic.claude-instant-v1100,000YesYes*
(*) The claude-3-5-sonnet models support streaming for tool use but may not adhere to the official OpenAI SDK standard for this feature.

​
Meta

Model IDMax TokensTool UseStreaming
meta.llama3-70b-instruct-v1:0128,000YesYes
meta.llama3-8b-instruct-v1:0128,000YesYes
us.meta.llama3-1-70b-instruct-v1:0128,000YesYes
us.meta.llama3-3-70b-instruct-v1:0128,000YesYes

​
Mistral

Model IDMax TokensTool UseStreaming
mistral.mistral-large-2402-v1:065,536YesYes
mistral.mixtral-8x7b-instruct-v0:132,768YesYes
mistral.mistral-small-2402-v1:032,768YesYes
mistral.mistral-7b-instruct-v0:232,768YesYes

​
Amazon

Model IDMax TokensTool UseStreaming
amazon.titan-text-premier-v1:030,720Check APIYes
amazon.titan-text-express-v18,192NoYes
amazon.titan-text-lite-v14,096NoYes
amazon.nova-micro-v1:02,048YesYes*

​
Cohere

Model IDMax TokensTool UseStreaming
cohere.command-r-plus-v1:0128,000YesYes
cohere.command-r-v1:0128,000YesYes
cohere.command-text-v148,192NoYes

​
AI121 Labs

Model IDMax TokensTool UseStreaming
ai21.jamba-instruct-v1:08,192YesYes
ai21.jamba-1-5-large-v1:08,192YesYes
ai21.jamba-1-5-mini-v1:08,192YesYes

​
DeepSeek

Model IDMax_TokensTool UseStreaming
us.deepseek.r1-v1:016,000Check APICheck API