Supported LLM Base Models

Navigator supports a variety of open source LLMs, and is adding more all the time. Below is the current list of supported models with links to their Hugging Face model pages where you can learn more about each model's strengths and weaknesses.

For most use cases that require training or inference on consumer hardware, we recommend using a 7B parameter model. This is the sweet spot between model performance and size for consumer devices. All LLM features in Navigator work well with 7B parameter models.

Full List of Supported Models

• Aya Models
  • Aya Expanse 8B 4bit
• CodeGemma Models
  • CodeGemma 7B IT 4bit
  • CodeGemma 7B IT 8bit
• Codestral Models
  • Codestral 22B v0.1 4bit
  • Codestral 22B v0.1 8bit
• DeepSeek Models
  • DeepSeek V3 0324
  • DeepSeek V3 0324 8bit
  • DeepSeek V3 0324 4bit
  • DeepSeek R1 Distill Llama 8B
  • DeepSeek R1 4bit
  • DeepSeek R1 0528 4bit
• Gemma Models
  • Gemma 7B IT
  • Gemma 2 2B IT
  • Gemma 2 9B IT
  • Gemma 2 27B IT
  • Gemma 3 1B IT 4bit
  • Gemma 3 1B IT 8bit
  • Gemma 3 1B IT 6bit
  • Gemma 3 1B IT bf16
  • Gemma 3 Text 4B IT 4bit
  • Gemma 3 Text 12B IT 4bit
  • Gemma 3 Text 27B IT 4bit
  • Gemma 3 4B IT 4bit
  • Gemma 3 4B IT bf16
  • Gemma 3 12B IT 4bit
  • Gemma 3 12B IT 8bit
  • Gemma 3 27B IT 4bit
  • Gemma 3 27B IT 8bit
  • Gemma 3 27B IT QAT bf16
  • Gemma 3n E2B IT 3bit
  • Gemma 3n E2B IT 4bit
  • Gemma 3n E2B IT 5bit
  • Gemma 3n E2B IT 6bit
  • Gemma 3n E2B IT 8bit
  • Gemma 3n E4B IT 3bit
  • Gemma 3n E4B IT 5bit
  • Gemma 3n E4B IT 6bit
  • Gemma 3n E4B IT 8bit
• Hunyuan Models
  • Hunyuan A52B Instruct 3bit
• Idefics Models
  • Idefics3 8B Llama3 3bit
  • Idefics3 8B Llama3 4bit
  • Idefics3 8B Llama3 6bit
  • Idefics3 8B Llama3 8bit
• Llama Models
  • Meta Llama 3 8B Instruct
  • Meta Llama 3 8B Instruct 4bit
  • Meta Llama 3 70B Instruct 4bit
  • Meta Llama 3 70B Instruct 8bit
  • Llama 3.1 8B Instruct
  • Meta Llama 3.1 8B Instruct
  • Meta Llama 3.1 8B Instruct 4bit
  • Meta Llama 3.1 8B Instruct 8bit
  • Meta Llama 3.1 8B Instruct bf16
  • Meta Llama 3.1 70B Instruct 4bit
  • Meta Llama 3.1 70B Instruct 8bit
  • Llama 3.1 405B Instruct
  • Llama 3.1 Nemotron 70B Instruct HF 4bit
  • Llama 3.2 1B Instruct
  • Llama 3.2 1B Instruct 4bit
  • Llama 3.2 1B Instruct 8bit
  • Llama 3.2 1B Instruct bf16
  • Llama 3.2 3B Instruct
  • Llama 3.2 3B Instruct 4bit
  • Llama 3.2 3B Instruct 8bit
  • Llama 3.2 3B Instruct bf16
  • Llama 3.3 70B Instruct
  • Llama 3.3 70B Instruct 4bit
  • Llama 3.3 70B Instruct 8bit
  • Llama 4 Scout 17B 16E Instruct
  • Llama 4 Scout 17B 16E Instruct 4bit
  • Llama 4 Scout 17B 16E Instruct 6bit
  • Llama 4 Scout 17B 16E Instruct 8bit
  • Llama 4 Scout 17B 16E Instruct bf16
  • Llama 4 Maverick 17B 128E Instruct 4bit
  • Llama 4 Maverick 17B 128E Instruct 6bit
  • Llama 3 Taiwan 70B Instruct
  • Llama 3.1 Nemotron Nano 8B v1
• Mamba Models
  • Mamba 1.4B HF f32
• Ministral Models
  • Ministral 8B Instruct 2410 bf16
• Mistral Models
  • Mistral 7B Instruct v0.3
  • Mistral 7B Instruct v0.3 (MLX)
  • Mistral 7B Instruct v0.3 4bit
  • Mistral 7B Instruct v0.3 8bit
  • Mistral NmbeddingMo Minitron 8B Instruct
  • Mistral Large Instruct 2407 4bit
  • Mistral Nemo Instruct 2407 4bit
  • Mistral Small 24B Instruct 2501
• Mixtral Models
  • Mixtral 8x7B Instruct v0.1
  • Mixtral 8x22B Instruct v0.1
  • Mixtral 8x22B Instruct v0.1 4bit
• Nemotron Models
  • Nemotron 4 340B Instruct HF
• PaliGemma Models
  • PaliGemma 3B PT 448
  • PaliGemma 3B Mix 224 8bit
  • PaliGemma 3B Mix 448 8bit
  • ColPaliGemma 3B PT 448 Base
• Phi Models
  • Phi 3 Mini 128K Instruct
  • Phi 3 Medium 128K Instruct bf16
  • Phi 3.5 Mini Instruct
  • Phi 3.5 Mini Instruct bf16
  • Phi 3.5 MoE Instruct
  • Phi 3.5 MoE Instruct 4bit
  • Phi 4
• Phixtral Models
  • Phixtral 2x2.8
  • Phixtral 2x2.8 4bit
• Qwen Models
  • Qwen2 7B Instruct
  • Qwen2 7B Instruct 8bit
  • Qwen2 7B Instruct 8bit (MLX)
  • Qwen2 72B Instruct 4bit
  • Qwen2.5 72B Instruct 4bit
  • Qwen2.5 14B Instruct
  • Qwen2.5 Coder 7B Instruct
  • QwQ 32B Preview
  • QwQ 32B
• Qwen3 Models
  • Qwen3 0.6B
  • Qwen3 0.6B 4bit
  • Qwen3 0.6B 6bit
  • Qwen3 0.6B 8bit
  • Qwen3 0.6B bf16
  • Qwen3 1.7B
  • Qwen3 1.7B 3bit
  • Qwen3 1.7B 4bit
  • Qwen3 1.7B 6bit
  • Qwen3 1.7B 8bit
  • Qwen3 1.7B bf16
  • Qwen3 4B
  • Qwen3 4B 3bit
  • Qwen3 4B 4bit
  • Qwen3 4B 6bit
  • Qwen3 4B 8bit
  • Qwen3 4B bf16
  • Qwen3 8B
  • Qwen3 8B 3bit
  • Qwen3 8B 4bit
  • Qwen3 8B 6bit
  • Qwen3 8B 8bit
  • Qwen3 30B A3B 4bit
  • Qwen3 30B A3B 6bit
  • Qwen3 30B A3B 8bit
  • Qwen3 30B A3B bf16
  • Qwen3 235B A22B 3bit
  • Qwen3 235B A22B 4bit
  • Qwen3 235B A22B 8bit
• Qwen VL Models
  • Qwen2.5 VL 7B Instruct 4bit
  • Qwen2.5 VL 7B Instruct 8bit
  • Qwen2.5 VL 32B Instruct 4bit
  • Qwen3 VL 2B Instruct 8bit
  • Qwen3 VL 2B Thinking 4bit
  • Qwen3 VL 2B Thinking bf16
  • Qwen3 VL 4B Instruct 8bit
  • Qwen3 VL 4B Thinking 4bit
  • Qwen3 VL 4B Thinking 8bit
  • Qwen3 VL 4B Thinking bf16
  • Qwen3 VL 8B Instruct 8bit
  • Qwen3 VL 8B Instruct bf16
  • Qwen3 VL 8B Thinking 4bit
  • Qwen3 VL 8B Thinking 8bit
  • Qwen3 VL 8B Thinking bf16
  • Qwen3 VL 30B A3B Instruct 4bit
  • Qwen3 VL 30B A3B Instruct 8bit
  • Qwen3 VL 30B A3B Instruct bf16
  • Qwen3 VL 30B A3B Thinking 4bit
  • Qwen3 VL 30B A3B Thinking 8bit
  • Qwen3 VL 30B A3B Thinking bf16
  • Qwen3 VL 32B Instruct 4bit
  • Qwen3 VL 32B Instruct 8bit
  • Qwen3 VL 32B Instruct bf16
  • Qwen3 VL 32B Thinking 4bit
  • Qwen3 VL 235B A22B Instruct 4bit
• SEA Models
  • Gemma SEA-LION v3 9B IT
  • Llama SEA-LION v2 8B IT
  • Llama SEA-LION v3 70B IT
  • Llama SEA-LION v3 8B IT
  • Llama SEA-LION v3.5 70B R
  • Llama SEA-LION v3.5 8B R
• SmolVLM Models
  • SmolVLM2 2.2B Instruct MLX
• Stable Code Models
  • Stable Code 3B 4bit
• Other Models
  • GPT-OSS 120B MLX 8bit
  • GPT-OSS 20B MLX 8bit
  • GPT-OSS 120B 4bit
  • Sum Small Unquantized

Model Selection Considerations

When selecting a model, consider these key factors:

• Parameter Count: Generally, larger models (higher parameter counts) offer better performance but require more computing resources.
• Memory Requirements: Each model requires a specific amount of RAM. Ensure your hardware meets these requirements.
• Specialization: Some models are optimized for specific tasks (e.g., code generation, instruction following).
• Quantization: Many models offer quantized versions (3-bit, 4-bit, 5-bit, 6-bit, 8-bit) that reduce memory requirements at a small cost to performance.