The AI Bench
24 GB · 153 GB/s

What can the MacBook Air (M5, 24 GB) run?

Silent and portable. Fine for small models; slow on big ones.

Best models for each task

Our top three for this machine. Chat and image tasks assume an 8K-token conversation; coding, documents and agents assume 32K.

Chat & writing18 of 25 run

  1. Gemma 4 12BRuns wellQ8_0 · needs 14.1 GB at 8K · 7–12tok/sReadable
  2. Qwen3.8 27BRuns wellUD-IQ3_S · needs 14.5 GB at 8K · 7–12tok/sReadable
  3. MiniCPM5 2BRuns greatQ8_0 · needs 4.0 GB at 8K · 30–50tok/sFast

Coding20 of 26 run

  1. Qwen3.5 4BRuns greatQ8_0 · needs 7.2 GB at 32K · 16–25tok/sSmooth
  2. Gemma 4 12BRuns greatQ4_K_M · needs 9.2 GB at 32K · 11–19tok/sSmooth
  3. gpt-oss 20BRuns wellMXFP4 · needs 13.4 GB at 32K · 25–50tok/sSmooth

Long documents13 of 20 run

  1. Qwen3.5 4BRuns greatQ8_0 · needs 7.2 GB at 32K · 16–25tok/sSmooth
  2. MiniCPM5 2BRuns greatQ8_0 · needs 5.1 GB at 32K · 20–35tok/sSmooth
  3. Gemma 4 12BRuns greatQ4_K_M · needs 9.2 GB at 32K · 11–19tok/sSmooth

Agents & tools20 of 26 run

  1. Qwen3.5 4BRuns greatQ8_0 · needs 7.2 GB at 32K · 16–25tok/sSmooth
  2. MiniCPM5 2BRuns greatQ8_0 · needs 5.1 GB at 32K · 20–35tok/sSmooth
  3. Ornith 1.5 9BRuns greatQ4_K_M · needs 7.9 GB at 32K · 13–20tok/sSmooth

Images in9 of 13 run

  1. Qwen3.8 27BRuns wellUD-IQ3_S · needs 14.5 GB at 8K · 7–12tok/sReadable
  2. Gemma 4 12BRuns wellQ8_0 · needs 14.1 GB at 8K · 7–12tok/sReadable
  3. Qwen3.5 9BRuns greatQ8_0 · needs 11.7 GB at 8K · 9–15tok/sReadable

Make images6 of 6 run

  1. Stable Diffusion XL Base 1.0Runs greatFP16 · needs 7.9 GB
  2. FLUX.2 [klein] 4BRuns wellBF16 · needs 14.0 GB
  3. Z-Image-TurboRuns great4-bit GGUF · needs 5.0 GB

Speak text7 of 7 run

  1. Qwen3-TTS 1.7BRuns greatBF16 · needs 5.5 GB
  2. Kokoro-82MRuns greatFP32 · needs 1.4 GB
  3. Chatterbox Multilingual V3Runs greatFull · needs 4.2 GB

Transcribe7 of 7 run

  1. Whisper large-v3-turbo (via faster-whisper)Runs greatFP16 · needs 2.6 GB
  2. Nemotron 3.5 ASR Streaming 0.6BRuns greatFP32 · needs 3.5 GB
  3. Qwen3-ASR 1.7BRuns greatBF16 · needs 5.7 GB

Biggest model it runs well

Qwen3.8 27B (UD-IQ3_S, 12.0 GB file) at about 7–12tok/sReadable.

Should you buy it?

Buy it if

You want local AI on a light laptop.

Skip it if

It will sit on a desk — a Mac mini with M5 Pro is twice as fast.

Watch out for

  • No fan: speed drops in long sessions.
  • 153 GB/s bandwidth limits speed.