Recommended models (10)

Qwen 3 0.6B

Alibaba · Apache 2.0

89% Match

Ultra-light Qwen 3 model for constrained devices

0.8 GBchat, edge

Llama 3.2 1B

Meta · Llama 3.2 Community

89% Match

Meta's smallest Llama for edge devices

1 GBchat, edge

Qwen 3.5 0.8B

Alibaba · Apache 2.0

86% Match

Ultra-tiny model for embedded and edge

0.9 GBchat, edge

Qwen 3 1.7B

Alibaba · Apache 2.0

80% Match

Compact Qwen 3 for mobile and edge

1.5 GBchat, edge

Phi-4 Mini 3.8B

Microsoft · MIT

76% Match

Microsoft's compact reasoning model

2.8 GBchat, code, reasoning

Gemma 3 4B

Google · Gemma

76% Match

Google's compact multimodal model

3 GBchat, vision

Llama 3.1 8B

Meta · Llama 3.1 Community

73% Match

Meta's versatile 8B — great quality/speed ratio

4.6 GBchat, code, reasoning

Qwen 3.5 9B

Alibaba · Apache 2.0

66% Match

Multimodal Qwen 3.5 mid-size

5.1 GBchat, vision

GPT-OSS 20B

OpenAI · Apache 2.0

53% Match

OpenAI's open-weight MoE with configurable reasoning

11.3 GBchat, reasoning, code

Mistral Small 3.1 24B

Mistral AI · Apache 2.0

45% Match

Multimodal Mistral with vision support

12.8 GBchat, vision, code