Available AIM containers#
CohereLabs#
Command A Reasoning 08 2025 (stable)
Cohere’s Command A Reasoning is a multilingual text generation model optimized for conversational AI and reasoning tasks across 23+ languages.
Qwen#
Qwen3 235B A22B (stable)
Qwen3 is the latest generation of LLMs in the Qwen series, offering a comprehensive suite models. This is the 235B parameter mixture–of–experts (MoE) variant.
Qwen3 32B (stable)
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture–of–experts (MoE) models.
#
Meta-llama#
Llama 3.1 405B Instruct (stable)
Massive instruction–tuned version of Llama 3.1 with 405B parameters for the most demanding tasks.
Llama 3.1 8B Instruct (stable)
Meta Llama 3.1 8B model optimized for chat and instruction following. Built on transformer architecture with GQA and RLHF training.
Llama 3.2 1B Instruct (stable)
Compact instruction–tuned Llama 3.2 model with 1B parameters for edge deployment.
Llama 3.2 3B Instruct (stable)
Balanced instruction–tuned Llama 3.2 model with 3B parameters.
Llama 3.3 70B Instruct (preview)
Instruction–tuned Llama 3.3 model with 70B parameters and improved performance.
Mistralai#
Mistral Small 3.2 24B Instruct 2506 (stable)
It is a minor update of Mistral–Small–3.1–24B–Instruct–2503 that improves instruction following and function calling robustness, and reduces repetition errors.
Mixtral 8x22B Instruct v0.1 (stable)
Mixture of experts model with 8 experts of 22B parameters each for efficient scaling.
Mixtral 8x7B Instruct v0.1 (stable)
Mixture of experts model with 8 experts of 7B parameters each for efficient scaling.
Openai#
GPT OSS 120B (stable)
OpenAI’s GPT–OSS 120B is a text generation model with conversational capabilities, supporting tool calling and reasoning tasks.
GPT OSS 20B (stable)
OpenAI’s GPT–OSS 20B is a text generation model with conversational capabilities, supporting tool calling and reasoning tasks.