What LLMs can MacBook Air M2 8 GB run?
8 GB unified, 100 GB/s memory bandwidth, about $1,199. Every open model in our catalog, rated by the engine for chat and for agentic coding.
Best for chat
Nothing in our catalog is usable for this.
Best for agentic coding
Nothing in our catalog is usable for this.
Best open-source (OSI) pick
Nothing in our catalog is usable for this.
Close your other apps and these will run
The ratings below assume a browser with a few tabs is open next to your chat app. With nothing else running, MacBook Air M2 8 GB has enough room for:
Doesn't fit or isn't supported (48)
Not enough usable memory, or the runtime can't run it here.
| Model | Chat | Agentic coding | Quant | Chat speed |
|---|---|---|---|---|
| DeepSeek-V4.1-Flash (748B-A16B)DeepSeekOpen source | Does Not Fit | Does Not Fit | — | — |
| Kolibri-1 (78B-A3.5B)Aleph AlphaOpen source | Does Not Fit | Does Not Fit | — | — |
| LFM2.5 24B-A2BLiquid AI | Does Not Fit | Does Not Fit | — | — |
| Muse Spark Open 816B-A42BMeta | Does Not Fit | Does Not Fit | — | — |
| Nemotron 3.5 Bolt 74B-A9BNVIDIA | Does Not Fit | Does Not Fit | — | — |
| Nemotron 3.5 Super VL 120B-A12BNVIDIA | Does Not Fit | Does Not Fit | — | — |
| Reflection Beam (501B-A23B)Reflection AIOpen source | Does Not Fit | Does Not Fit | — | — |
| Step 5 Preview 600B-A27BStepfun | Does Not Fit | Does Not Fit | — | — |
| MiniCPM5 2BOpenBMBOpen source | Does Not Fit | Does Not Fit | — | — |
| GLM-5.3 (753B-A40B)Z.ai | Does Not Fit | Does Not Fit | — | — |
| GLM-5.3-Flash (320B-A18B)Z.aiOpen source | Does Not Fit | Does Not Fit | — | — |
| Muse Glimmer (30B)MetaOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.8 27BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Gemma 4 12BGoogleOpen source | Does Not Fit | Does Not Fit | — | — |
| DeepSeek-V4-Flash (284B-A13B)DeepSeekOpen source | Does Not Fit | Does Not Fit | — | — |
| Gemma 4 26B-A4BGoogleOpen source | Does Not Fit | Does Not Fit | — | — |
| Gemma 4 31BGoogleOpen source | Does Not Fit | Does Not Fit | — | — |
| Gemma 4 E4BGoogleOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.6 27BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.6 35B-A3BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Mistral Small 4 (119B-A6.5B)Mistral AIOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.5 122B-A10BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.5 27BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.5 35B-A3BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.5 4BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3.5 9BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| GLM-4.7-Flash (30B-A3B)Z.aiOpen source | Does Not Fit | Does Not Fit | — | — |
| Devstral Small 2 (24B)Mistral AIOpen source | Does Not Fit | Does Not Fit | — | — |
| GLM-4.6V-Flash (10B)Z.aiOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3-Next 80B-A3B InstructAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| gpt-oss-120bOpenAIOpen source | Does Not Fit | Does Not Fit | — | — |
| gpt-oss-20bOpenAIOpen source | Does Not Fit | Does Not Fit | — | — |
| GLM-4.5-Air (106B-A12B)Z.aiOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3 30B-A3B Instruct 2507Alibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3-Coder 30B-A3B InstructAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Mistral Small 3.2 (24B)Mistral AIOpen source | Does Not Fit | Does Not Fit | — | — |
| DeepSeek-R1-0528-Qwen3-8BDeepSeekOpen source | Does Not Fit | Does Not Fit | — | — |
| Llama 4 Scout (109B-A17B)MetaOpen weights (restricted) | Does Not Fit | Does Not Fit | — | — |
| Qwen3 14BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3 32BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen3 8BAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Gemma 3 27BGoogleOpen weights (restricted) | Does Not Fit | Does Not Fit | — | — |
| Llama 3.3 70B InstructMetaOpen weights (restricted) | Does Not Fit | Does Not Fit | — | — |
| Phi-4 (14B)MicrosoftOpen source | Does Not Fit | Does Not Fit | — | — |
| Qwen2.5-Coder 32B InstructAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Llama 3.2 3B InstructMetaOpen weights (restricted) | Does Not Fit | Does Not Fit | — | — |
| Qwen2.5-Coder 7B InstructAlibaba QwenOpen source | Does Not Fit | Does Not Fit | — | — |
| Llama 3.1 8B InstructMetaOpen weights (restricted) | Does Not Fit | Does Not Fit | — | — |
Want a specific tool, context size or workload?
The full check models your exact setup, including the apps you keep open.