Available AIM containers#
Instinct#
Cohere Labs#
111B parameter language model with configurable reasoning and tool use capabilities.
DeepSeek#
deepseek-ai/DeepSeek-R1 (stable)
671B parameter MoE reasoning model with 37B active parameters.
deepseek-ai/DeepSeek-R1-0528 (stable)
671B parameter MoE reasoning model with 37B active parameters, updated version of DeepSeek-R1.
deepseek-ai/DeepSeek-V3.1 (stable)
671B parameter MoE model with 37B active parameters supporting thinking and non-thinking modes.
deepseek-ai/DeepSeek-V3.1-Terminus (stable)
671B parameter MoE model with 37B active parameters, refined for language consistency and agent tasks.
Google#
google/gemma-3-1b-it (stable)
Gemma 3 1B IT is a lightweight instruction-tuned model supporting text generation with a 32K context window.
google/gemma-3-27b-it (stable)
Gemma 3 27B IT is a multimodal instruction-tuned model supporting text and image input with a 128K context window.
google/gemma-4-31B-it (preview)
Gemma 4 31B IT is a multimodal instruction-tuned model with text and image input, 256K native context, and Gemma 4 reasoning + tool-call parsers.
google/medgemma-27b-it (stable)
Gemma 3-based 27B multimodal model fine-tuned for medical text and image tasks (X-ray, dermatology, ophthalmology, pathology, radiology reports).
Meta#
meta-llama/Llama-3.1-405B-Instruct (stable)
Multilingual 405B parameter instruction-tuned language model for dialogue use cases.
meta-llama/Llama-3.1-8B-Instruct (stable)
Multilingual 8B parameter instruction-tuned language model for dialogue use cases.
meta-llama/Llama-3.2-1B-Instruct (stable)
Multilingual 1B parameter instruction-tuned language model for dialogue and on-device use cases.
meta-llama/Llama-3.2-3B-Instruct (stable)
Multilingual 3B parameter instruction-tuned language model for dialogue and on-device use cases.
meta-llama/Llama-3.3-70B-Instruct (stable)
Multilingual 70B parameter instruction-tuned language model for dialogue use cases.
MiniMax#
MiniMaxAI/MiniMax-M2.5 (stable)
228B parameter mixture-of-experts language model with reasoning, tool calling, and coding capabilities.
Mistral AI#
14B parameter instruction-tuned language model with vision and function calling capabilities.
14B parameter instruction-tuned language model with vision and function calling capabilities.
675B parameter granular MoE multimodal model with 41B active parameters and vision capabilities.
24B parameter instruction-tuned language model (Mistral Small 3) with native function calling. Text-only.
24B parameter instruction-tuned language model with vision and function calling capabilities.
Sparse MoE language model with 141B total parameters across 8 experts and function calling support.
mistralai/Mixtral-8x7B-Instruct-v0.1 (stable)
Sparse MoE language model with 47B total parameters across 8 experts.
OpenAI#
openai/gpt-oss-120b (stable)
Open-weight 117B parameter MoE model with 5.1B active parameters and configurable reasoning.
openai/gpt-oss-20b (stable)
Open-weight 21B parameter MoE model with 3.6B active parameters for lower-latency use cases.
Qwen#
Qwen/Qwen3-235B-A22B (stable)
235B parameter MoE language model with 22B active parameters and dual thinking modes.
Qwen/Qwen3-32B (stable)
32.8B parameter dense language model with dual thinking modes and multilingual support.
Qwen/Qwen3-Coder-Next (stable)
80B parameter MoE coding agent model with 3B active parameters and hybrid attention architecture.
Qwen/Qwen3-VL-235B-A22B-Instruct (stable)
236B parameter MoE vision-language model with 22B active parameters and multimodal capabilities.
Qwen/Qwen3-VL-235B-A22B-Thinking (stable)
236B parameter MoE vision-language model with reasoning-enhanced thinking capabilities.
Z.ai#
zai-org/GLM-4.7 (stable)
GLM-4.7 is a large language model with multi-turn conversation, tool use, and reasoning capabilities.
EPYC#
Google#
google/gemma-4-E4B-it (stable)
Gemma 4 E4B is Google’s efficient instruction-tuned model with a ~4B effective-parameter MatFormer architecture for low-latency on-device and CPU inference.
Meta#
meta-llama/Llama-3.1-8B-Instruct (stable)
Multilingual 8B parameter instruction-tuned language model for dialogue use cases.
meta-llama/Llama-3.2-1B-Instruct (preview)
Multilingual 1B parameter instruction-tuned language model for dialogue and on-device use cases.
meta-llama/Llama-3.2-3B-Instruct (preview)
Multilingual 3B parameter instruction-tuned language model for dialogue and on-device use cases.
Qwen#
Qwen/Qwen3-30B-A3B (stable)
Mixture-of-experts 30B (3B active) LLM with thinking/non-thinking mode switching, advanced reasoning, agent capabilities, and 100+ language support.
Qwen/Qwen3-8B (stable)
Reasoning-enhanced 8B parameter LLM with thinking/non-thinking mode switching, excelling in math, coding, and multi-turn conversations.
Qwen/Qwen3.5-4B (stable)
Qwen3.5-4B is a 4B parameter LLM with thinking/non-thinking dual-mode reasoning, strong math and coding ability, and multilingual support.
Qwen/Qwen3.5-9B (stable)
Qwen3.5-9B is a 9B parameter LLM with thinking/non-thinking dual-mode reasoning, strong math and coding ability, and multilingual support.
Qwen/Qwen3.6-35B-A3B (stable)
Qwen3.6-35B-A3B is a Mixture-of-Experts LLM with 35B total parameters and ~3B active per token, balancing high quality with efficient inference.
unsloth#
unsloth/gpt-oss-20b-BF16 (stable)
gpt-oss-20B is OpenAI’s open-weight 20B Mixture-of-Experts model (BF16 conversion by Unsloth) with strong reasoning and tool-use capabilities.
Radeon#
Google#
google/gemma-3n-E4B-it (preview)
Gemma 3n E4B IT is a gated multimodal instruction-tuned model supporting text, image, video, and audio inputs.
Meta#
meta-llama/Llama-3.1-8B-Instruct (preview)
Multilingual 8B parameter instruction-tuned language model for dialogue use cases.
Qwen#
Qwen/Qwen3-VL-8B-Instruct (preview)
8B parameter vision-language model with advanced multimodal reasoning.
Qwen/Qwen3.5-9B (preview)
9B parameter hybrid language model with Gated DeltaNet and dual thinking modes.
Z.ai#
zai-org/GLM-4.7-Flash (preview)
30B-A3B MoE language model with balanced performance and efficiency for lightweight deployment.