MacBook Air 13-inch M4 · 16 GB unified
Apple · MacBook Air 13-inch · 2025
594 text models run well · 120 GB/s bandwidth
Models on this hardware
Start here
File Q4_K_M · 0.8 GB · bartowski/Llama-3.2-1B-Instruct-GGUF · What is Q4?
Memory Needs ~1 GB · ~5 GB on MacBook Air 13-inch 2025 · M4 · 16 GB unified, including ~4 GB the system keeps
ollama run hf.co/bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M- Install Ollama
- Run the command above
- Ask:
Explain what a context window is, in two sentences.
File Q4_K_M · 2.2 GB · ibm-granite/granite-4.2-3b-GGUF · What is Q4?
Memory Needs ~3 GB · ~7 GB on MacBook Air 13-inch 2025 · M4 · 16 GB unified, including ~4 GB the system keeps
ollama run hf.co/ibm-granite/granite-4.2-3b-GGUF:Q4_K_M- Install Ollama
- Run the command above
- Ask:
Explain what a context window is, in two sentences.
File Q4_K_M · 5.3 GB · ibm-granite/granite-4.2-8b-GGUF · What is Q4?
Memory Needs ~6 GB · ~10 GB on MacBook Air 13-inch 2025 · M4 · 16 GB unified, including ~4 GB the system keeps
ollama run hf.co/ibm-granite/granite-4.2-8b-GGUF:Q4_K_M- Install Ollama
- Run the command above
- Ask:
Explain what a context window is, in two sentences.
Text
Text · LMArena
rating · higher is better
- Color = publisher
Source: LMArena 2026-10-22 · collected 2026-10-25 · dataset · CC BY 4.0 · Models without a rating are left out, not scored as zero.
| Granite 4.2 3BQ4 · published params · source | 1,290 ± 12 | ~7 GB | Q4 · published params |
|---|---|---|---|
| Granite 4.2 8BQ4 · published params · source | 1,290 ± 11 | ~10 GB | Q4 · published params |
| Phi-4Q4 · published params · source | 1,256 ± 5 | ~15 GBTight fit: Needs ~15 GB: runs well, about 1 GB left beyond the system reserve | Q4 · published params |
| Llama 3.1 8B InstructQ4 · published params · source | 1,211 ± 4 | ~10 GB | Q4 · published params |
| Llama 3.2 3B InstructQ4 · published params · source | 1,167 ± 8 | ~7 GB | Q4 · published params |
| Llama 3.2 1B InstructQ4 · published params · source | 1,111 ± 8 | ~5 GB | Q4 · published params |
| Qwen3.5 9BQ4 · published params · source | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | Q4 · published params |
| Qwen3 8BQ4 · published params · source | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | Q4 · published params |
No model within 16 GB has a LiveBench score, so the table has no LiveBench column. LMArena ratings in this table: LMArena 2026-10-22 · collected 2026-10-25 · dataset · CC BY 4.0
Tight fit: it runs, but leaves less than 1.6 GB free beyond the 4 GB the system keeps (the memory figure already includes that reserve).
Every reviewed text model here 21
Every one that runs here, well or slowly, scored or not; 6 of 21 have a LiveBench or LMArena result. Run well first, then tight fits, then slow; within each, by Hub downloads over 30 days, read 2026-11-19, not by quality.
| Qwen3.5 9B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | 5.7 | Apache 2.0 | 6.5M |
|---|---|---|---|---|---|---|
| Qwen3 8B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | 5.0 | Apache 2.0 | 6.1M |
| Gemma 4 E4B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~10 GB | 5.0 | Apache 2.0 | 4.4M |
| Llama 3.1 8B Instruct | Not in LiveBench 2026-06-25 | 1,211 | ~10 GB | 4.9 | Llama 3.1 | 4.1M |
| Gemma 4 E2B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~8 GB | 3.1 | Apache 2.0 | 2.8M |
| Qwen3 4B Instruct 2507 | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~7 GB | 2.5 | Apache 2.0 | 2.8M |
| Qwen3 1.7B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~6 GB | 1.1 | Apache 2.0 | 2.7M |
| Gemma 4 12B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~13 GB | 7.1 | Apache 2.0 | 1.6M |
| Llama 3.2 1B Instruct | Not in LiveBench 2026-06-25 | 1,111 | ~5 GB | 0.8 | Llama 3.2 | 1.3M |
| Llama 3.2 3B Instruct | Not in LiveBench 2026-06-25 | 1,167 | ~7 GB | 2.0 | Llama 3.2 | 823.8K |
| DeepSeek R1 0528 Qwen3 8B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | 5.0 | MIT | 523.8K |
| Phi-4 mini | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~7 GB | 2.5 | MIT | 301.1K |
| SmolLM3 3B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~7 GB | 1.9 | Apache 2.0 | 185.3K |
| Olmo 3 7B Instruct | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~10 GB | 4.5 | Apache 2.0 | 133.1K |
| Ministral 3 8B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~11 GB | 5.2 | Apache 2.0 | 90.7K |
| Ministral 3 3B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~7 GB | 2.1 | Apache 2.0 | 72.5K |
| Granite 4.2 8B | Not in LiveBench 2026-06-25 | 1,290 | ~10 GB | 5.3 | Apache 2.0 | 59.5K |
| Granite 4.2 3B | Not in LiveBench 2026-06-25 | 1,290 | ~7 GB | 2.2 | Apache 2.0 | 40.8K |
| Qwen3 14B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~16 GBTight fit | 9.0 | Apache 2.0 | 13.2M |
| Ministral 3 14B | Not in LiveBench 2026-06-25 | Not on the LMArena text leaderboard of 2026-10-22 | ~15 GBTight fit | 8.2 | Apache 2.0 | 207.6K |
| Phi-4 | Not in LiveBench 2026-06-25 | 1,256 | ~15 GBTight fit | 8.0 | MIT | 199.5K |
With 24 GB you also run… 2
The next memory size on this ladder runs these well. Figures are the requirement on this hardware type.
- gpt-oss-20bOpenAI~20 GBQ4 · published params
- Devstral Small 2 24BMistral~22 GBQ4 · published params
Adjust the estimate: quantization, context, reserve
Vision
Vision · LMArena
rating · higher is better
- Color = publisher
Source: LMArena 2026-10-19 · collected 2026-10-25 · dataset · CC BY 4.0 · Models without a rating are left out, not scored as zero.
| Molmo2-8BQ4 · Hugging Face count · source | 1,112 ± 20 | ~11 GB | Q4 · Hugging Face count |
|---|---|---|---|
| Qwen2-VL-7B-InstructQ4 · Hugging Face count · source | 1,031 ± 11 | ~11 GB | Q4 · Hugging Face count |
| Molmo-7B-D-0924Q4 · Hugging Face count · source | 994 ± 14 | ~11 GB | Q4 · Hugging Face count |
| Llama-3.2-11B-Vision-InstructQ4 · Hugging Face count · source | 993 ± 11 | ~13 GB | Q4 · Hugging Face count |
| MiniCPM-V-2_6Q4 · Hugging Face count · source | 962 ± 16 | ~11 GB | Q4 · Hugging Face count |
| Phi-3.5-vision-instructQ4 · Hugging Face count · source | 919 ± 16 | ~8 GB | Q4 · Hugging Face count |
| Gemma 4 12BQ4 · published params · source | Not on the LMArena vision leaderboard of 2026-10-19 | ~13 GB | Q4 · published params |
| Qwen3.5 9BQ4 · published params · source | Not on the LMArena vision leaderboard of 2026-10-19 | ~11 GB | Q4 · published params |
LMArena ratings in this table: LMArena 2026-10-19 · collected 2026-10-25 · dataset · CC BY 4.0
Tight fit: it runs, but leaves less than 1.6 GB free beyond the 4 GB the system keeps (the memory figure already includes that reserve).
With 24 GB you also run… 8
The next memory size on this ladder runs these well. Figures are the requirement on this hardware type.
- deepseek-vl2-smalldeepseek-ai~17 GBQ4 · Hugging Face count
- Kimi-VL-A3B-Instructmoonshotai~17 GBQ4 · Hugging Face count
- Kimi-VL-A3B-Thinkingmoonshotai~17 GBQ4 · Hugging Face count
- Kimi-VL-A3B-Thinking-2506moonshotai~17 GBQ4 · Hugging Face count
- LLaDA-UIinclusionAI~17 GBQ4 · Hugging Face count
- InternVideo2-Stage2_6BOpenGVLab~18 GB16-bit est. · Hugging Face count
- InternVL3_5-GPT-OSS-20B-A4B-Preview-HFOpenGVLab~20 GBQ4 · Hugging Face count
- RMBG-3.0briaai~22 GB16-bit est. · Hugging Face count
471 more vision models in the catalog have no usable memory figure or parameter count for this profile. Browse Vision models
Image
Image · Prompt alignment · GenEval
0–1 · higher is better
Source: GenEval · 2023 · reviewed 2026-10-05 · Published quality in the original benchmark preset; not measured on this hardware.
| Stable Diffusion XL 1.0Platform path documented · Apple Core ML · source | 0.55 | Not on the LMArena text-to-image leaderboard of 2026-10-07 | Documented path without a memory figure | Platform path documented |
|---|---|---|---|---|
| Stable Diffusion 1.5Platform path documented · Apple Core ML · source | 0.43 | Not on the LMArena text-to-image leaderboard of 2026-10-07 | Documented path without a memory figure | Platform path documented |
| flux-3-image-klein16-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~16 GB | 16-bit est. · params |
| flux-3-image-klein-base16-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~16 GB | 16-bit est. · params |
| MOSS-Image-Base16-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~15 GB | 16-bit est. · params |
| MOSS-Image-SFT16-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~15 GB | 16-bit est. · params |
| ideogram-4-nf416-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~14 GB | 16-bit est. · params |
| ideogram-4-nf4-diffusers16-bit est. · params · source | No comparable published result | Not reviewed for LMArena | ~14 GB | 16-bit est. · params |
LMArena ratings in this table: LMArena 2026-10-07 · collected 2026-10-25 · dataset · CC BY 4.0
With 24 GB you also run… 37
The next memory size on this ladder. Figures are the requirement on this hardware type.
- Kandinsky-5.0-I2I-Lite-pretrain-Diffuserskandinskylab~17 GB16-bit est. · params
- Kandinsky-5.0-I2I-Lite-sft-Diffuserskandinskylab~17 GB16-bit est. · params
- Kandinsky-5.0-T2I-Lite-pretrain-Diffuserskandinskylab~17 GB16-bit est. · params
- Kandinsky-5.0-T2I-Lite-sft-Diffuserskandinskylab~17 GB16-bit est. · params
- LongCat ImageMeituan~17 GB16-bit est. · params
- Ming-Image-0.1-DesigninclusionAI~17 GB16-bit est. · params
- Ming-Image-0.1-Design-LayerinclusionAI~17 GB16-bit est. · params
- Z-ImageTongyi-MAI~17 GB16-bit est. · params
- CogView4-6Bzai-org~18 GB16-bit est. · params
- LLaDA-ImageinclusionAI~18 GB16-bit est. · params
- LLaDA-Image-TurboinclusionAI~18 GB16-bit est. · params
- Janus ProDeepSeek~19 GB16-bit est. · params
- OmniGen V2VectorSpaceLab~19 GB16-bit est. · sum of parts
- Qwen-Image-2.1Qwen~20 GB16-bit est. · params
- Qwen-Image-2.1-TurboQwen~20 GB16-bit est. · params
- ERNIE ImageBaidu~22 GB16-bit est. · params
- ERNIE Image TurboBaidu~22 GB16-bit est. · params
- FIBOBria~22 GB16-bit est. · params
- FIBO LiteBria~22 GB16-bit est. · params
- Fibo-1.5briaai~22 GB16-bit est. · params
- Fibo-2.0-devbriaai~22 GB16-bit est. · params
- Fibo-2.0-edit-turbobriaai~22 GB16-bit est. · params
- Fibo-Editbriaai~22 GB16-bit est. · params
- Fibo-Edit-1.5-basebriaai~22 GB16-bit est. · params
- Fibo-Edit-1.5-turbobriaai~22 GB16-bit est. · params
- Fibo-Edit-RMBGbriaai~22 GB16-bit est. · params
- HiDream-O1-ImageHiDream~22 GB16-bit est. · params
- HiDream-O1-Image-DevHiDream~22 GB16-bit est. · params
- HY-WUtencent~22 GB16-bit est. · params
- Infinity 8BByteDance~22 GB16-bit est. · params
- Stable Diffusion 3.5 LargeStability AI~22 GB16-bit est. · params
- Stable Diffusion 3.5 Large TurboStability AI~22 GB16-bit est. · params
- ETCHR-FLUX.2-klein-9Binternlm~24 GB16-bit est. · params
- FLUX.2 [klein] 9BBlack Forest Labs~24 GB16-bit est. · params
- FLUX.2 [klein] Base 9BBlack Forest Labs~24 GB16-bit est. · params
- FLUX.2-klein-9b-kvblack-forest-labs~24 GB16-bit est. · params
- HiDream-O1-Image-Dev-2604HiDream-ai~24 GB16-bit est. · params
231 more image models in the catalog have no usable memory figure or parameter count for this profile. Browse Image models
Video
| CogVideoX-5b-I2V16-bit est. · params · source | ~16 GB | 16-bit est. · params |
|---|---|---|
| CogVideoX1.5-5B16-bit est. · params · source | ~16 GB | 16-bit est. · params |
| CogVideoX1.5-5B-I2V16-bit est. · params · source | ~16 GB | 16-bit est. · params |
| SkyReels-A116-bit est. · params · source | ~16 GB | 16-bit est. · params |
| CogVideoX-5B16-bit est. · params · source | ~14 GB | 16-bit est. · params |
| EdenLive-5B-Causal-Diffusers16-bit est. · params · source | ~14 GB | 16-bit est. · params |
| EdenLive-5B-Diffusers16-bit est. · params · source | ~14 GB | 16-bit est. · params |
| Wan2.2-TI2V-5B16-bit est. · params · source | ~14 GB | 16-bit est. · params |
How this list works Documented execution paths on this hardware first, then publisher minimums (CUDA or Apple MPS).
With 24 GB you also run… 2
The next memory size on this ladder. Figures are the requirement on this hardware type.
- ContentV-8BByteDance~22 GB16-bit est. · params
- HunyuanVideo-1.5tencent~22 GB16-bit est. · params
125 more video models in the catalog have no usable memory figure or parameter count for this profile. Browse Video models
Audio
Audio · Word error rate · Open ASR
% · lower is better
Source: Word error rate · Open ASR · reviewed 2026-10-20 · Published quality in the original benchmark preset; not measured on this hardware.
| Whisper large-v3-turboPlatform path documented · whisper.cpp / Metal · source | 7.83 | Documented path without a memory figure | Platform path documented |
|---|---|---|---|
| Whisper large-v3Platform path documented · whisper.cpp / Metal · source | 7.44 | Documented path without a memory figure | Platform path documented |
| GaussianMind16-bit est. · params · source | No comparable published result | ~16 GB | 16-bit est. · params |
| higgs-tts-2-3b-base16-bit est. · params · source | No comparable published result | ~16 GB | 16-bit est. · params |
| paza-Phi-4-multimodal-instruct16-bit est. · params · source | No comparable published result | ~16 GB | 16-bit est. · params |
| Phi-4-multimodal-instruct16-bit est. · params · source | No comparable published result | ~16 GB | 16-bit est. · params |
| MOSS-Audio-4B-Instruct16-bit est. · params · source | No comparable published result | ~15 GB | 16-bit est. · params |
| MOSS-Audio-4B-Thinking16-bit est. · params · source | No comparable published result | ~15 GB | 16-bit est. · params |
With 24 GB you also run… 53
The next memory size on this ladder. Figures are the requirement on this hardware type.
- YuE-s1-7B-anneal-en-iclHKUSTAudio~17 GB16-bit est. · params
- YuE-s1-7B-anneal-jp-kr-cotHKUSTAudio~17 GB16-bit est. · params
- omniASR-W2V-7Bfacebook~18 GB16-bit est. · params
- drama-3-int4fishaudio~19 GB16-bit est. · params
- MiMo-V2.5-ASRXiaomiMiMo~21 GB16-bit est. · params
- moshika-candle-bf16kyutai~21 GB16-bit est. · params
- moshika-mlx-bf16kyutai~21 GB16-bit est. · params
- moshika-pytorch-bf16kyutai~21 GB16-bit est. · params
- moshika-rag-pytorch-bf16kyutai~21 GB16-bit est. · params
- moshika-rl-seamlesskyutai~21 GB16-bit est. · params
- moshiko-candle-bf16kyutai~21 GB16-bit est. · params
- moshiko-mlx-bf16kyutai~21 GB16-bit est. · params
- moshiko-pytorch-bf16kyutai~21 GB16-bit est. · params
- SongPrep-7Btencent~21 GB16-bit est. · params
- audio-flamingo-3-hfnvidia~22 GB16-bit est. · params
- audio-flamingo-next-captioner-hfnvidia~22 GB16-bit est. · params
- audio-flamingo-next-hfnvidia~22 GB16-bit est. · params
- audio-flamingo-next-think-hfnvidia~22 GB16-bit est. · params
- music-flamingo-2601-hfnvidia~22 GB16-bit est. · params
- music-flamingo-hfnvidia~22 GB16-bit est. · params
- music-flamingo-think-2601-hfnvidia~22 GB16-bit est. · params
- ultravox-v0_2fixie-ai~22 GB16-bit est. · params
- ultravox-v0_3fixie-ai~22 GB16-bit est. · params
- ultravox-v0_4fixie-ai~22 GB16-bit est. · params
- Unified_Audio_Schematencent~22 GB16-bit est. · params
- VibeVoice-ASR-HFmicrosoft~22 GB16-bit est. · params
- Covo-Audio-Chattencent~23 GB16-bit est. · params
- granite-speech-3.2-8bibm-granite~23 GB16-bit est. · params
- granite-speech-3.3-8bibm-granite~23 GB16-bit est. · params
- Llasa-8BHKUSTAudio~23 GB16-bit est. · params
- moshika-rag-candle-bf16kyutai~23 GB16-bit est. · params
- MOSS-SoundEffectOpenMOSS-Team~23 GB16-bit est. · params
- MOSS-TTSOpenMOSS-Team~23 GB16-bit est. · params
- MOSS-TTS-v1.5OpenMOSS-Team~23 GB16-bit est. · params
- moss-tts-v1.5-8b-voice-actinglaion~23 GB16-bit est. · params
- MOSS-TTSD-v1.0OpenMOSS-Team~23 GB16-bit est. · params
- personaplex-7b-v1nvidia~23 GB16-bit est. · params
- personaplex-rl-seamlesskyutai~23 GB16-bit est. · params
- Qwen2-Audio-7BQwen~23 GB16-bit est. · params
- Qwen2-Audio-7B-InstructQwen~23 GB16-bit est. · params
- VibeVoice-ASRmicrosoft~23 GB16-bit est. · params
- VibeVoice-ASR-Streaming-7Bmicrosoft~23 GB16-bit est. · params
- higgs-audio-v3-8b-sttbosonai~24 GB16-bit est. · params
- higgs-audio-v3-8b-stt-v2bosonai~24 GB16-bit est. · params
- Ming-3.0-asr-tinyinclusionAI~24 GB16-bit est. · params
- Ming-3.0-audio-tinyinclusionAI~24 GB16-bit est. · params
- Ming-3.0-tts-tinyinclusionAI~24 GB16-bit est. · params
- MOSS-Audio-8B-InstructOpenMOSS-Team~24 GB16-bit est. · params
- MOSS-Audio-8B-ThinkingOpenMOSS-Team~24 GB16-bit est. · params
- MOSS-Music-8B-InstructOpenMOSS-Team~24 GB16-bit est. · params
- MOSS-Music-8B-ThinkingOpenMOSS-Team~24 GB16-bit est. · params
- moss-vocal-burst-captionerlaion~24 GB16-bit est. · params
- moss-voice-captionerlaion~24 GB16-bit est. · params
1,986 more audio models in the catalog have no usable memory figure or parameter count for this profile. Browse Audio models
Embeddings
Embeddings · MTEB (Multilingual, v2) · mean(task)
0–100 · higher is better
- Color = publisher
| Qwen3-Embedding 8B8-bit · published params · source | 70.6 | ~12 GB | 8-bit · published params |
|---|---|---|---|
| Qwen3-Embedding 4B8-bit · published params · source | 69.5 | ~7 GB | 8-bit · published params |
| Harrier OSS v1 (0.6B)8-bit · published params · source | 69.0 | ~3 GB | 8-bit · published params |
| Qwen3-Embedding 0.6B8-bit · published params · source | 64.3 | ~3 GB | 8-bit · published params |
| Multilingual E5 Large Instruct8-bit · published params · source | 63.2 | ~3 GB | 8-bit · published params |
| EmbeddingGemma 28-bit · published params · source | 61.4 | ~3 GB | 8-bit · published params |
| EmbeddingGemma (300M)8-bit · published params · source | 61.1 | ~3 GB | 8-bit · published params |
| BGE-M38-bit · published params · source | 59.6 | ~3 GB | 8-bit · published params |
With 24 GB you also run… 2
The next memory size on this ladder runs these well. Figures are the requirement on this hardware type.
- KaLM-Embedding-Gemma3-12B-2511tencent~17 GB8-bit · Hub count
- InternVL-14B-224pxOpenGVLab~19 GB8-bit · Hub count
Runs well within 16 GB of unified memory · planning estimates, not measured speed
Measurements
Speed on this hardware
We have not yet included a verifiable measurement for this exact configuration. The memory estimates above do not indicate tokens per second or time to first token.Specifications
20 sourced fields; the source is named once per group, and a field from another source links to it. Official specifications
Marker: the lightest text model that does not run well here but runs well with 24 GB. No text model that runs well here has a LiveBench score. Q4 planning estimate, not a measured fit.
Selected configuration
Latest check · Manufacturer source
- Memory capacity
- 16 GB
- Memory type
- 16 GB unified
- Chip / processor
- Apple M4 · 10-core CPU · 8- or 10-core GPU
- Memory bandwidth
- 120 GB/s
- CPU
- Apple M4
- CPU cores
- 10
Family specification
Latest check · Manufacturer source
- Introduced
- 2025
- Storage options (family)
- 256 GB · 512 GB · 1 TB · 2 TB SSD
- Ports
- 2 × Thunderbolt 4 · MagSafe 3 · 3.5 mm
- Battery capacity
- 53.8 Wh
- Wi-Fi
- Wi-Fi 6E (802.11ax)
- Bluetooth
- Bluetooth 5.3
- Dimensions (cm)
- 30.41 × 21.5 × 1.13 (W × D × H)
- Display size
- 13.6 in
- Display resolution
- 2560 × 1664
- Neural Engine cores
- 16
- Maximum Thunderbolt rate
- 40 Gb/s
- Weight
- 1.24 kg
- Operating system
- macOS
The 13-inch MacBook Air with M4 starts at 16 GB unified memory. Apple describes MacBook Air as its most popular laptop, a family-level claim rather than sales evidence for this exact memory or GPU configuration. Apple lists 8- or 10-core GPU options, so this profile tracks memory capacity and does not imply a single GPU-core configuration.
Photo: Official manufacturer image. · Manufacturer source
Memory bandwidth is not measured model speed; verify runtime and checkpoint compatibility. Hardware specification reviewed: 2026-10-09.
How estimates work
- Candidates
- Reviewed pages use published parameter counts; automatic text candidates use official repository tensor counts (packed quantized weights and unsupported architectures excluded).
- Planning
- Q4 weights + 20% allowance + reserve: 25% of unified or system memory (at least 2 GB, at most 8 GB; 4 GB here) or 2 GB per dedicated GPU. Adjustable in “Adjust the estimate”. No KV cache is calculated from context.
- Fit tiers
- Runs well: the estimate fits the fast memory (VRAM, or unified memory within the macOS GPU share). Runs slowly: within the machine's memory but outside the fast part; a graphics card offloads layers to PC RAM (the listed RAM, or 32 GB assumed). Doesn't run: above both. Tiers are planning estimates, never a speed.
- Vision
- Vision-language models use the Q4 planning estimate from the parameter count (published by the model page, or read from Hugging Face). Classic vision models (classify, detect, segment) use a 16-bit estimate. No vision benchmark is tracked, so rows carry no score.
- Media and embeddings
- Media rows use documented execution paths (exact hardware or platform) and publisher minimums; a published requirement above this memory moves a model to the next tier. Without either, a 16-bit estimate from the published parameter count (× 1.2 + 2 GB), labelled as such. Embeddings use 8-bit weights + 20% + 2 GB.
Local AI basics: parameters, Q4 and memory · Full model catalog · Data and methodology
Alternatives
Other machines on the same memory ladder (unified memory); each page lists the models that fit it.
Also 16 GB 8
- iMac 24-inch · M4 · 8-core CPU / 8-core GPU · 16 GB unified
- Mac mini M1 · 16 GB unified
- Mac mini M2 · 16 GB unified
- Mac mini M4 · 16 GB unified
- Mac mini M6 · 16 GB unified
- MacBook Air 13-inch M5 · 16 GB unified
- MacBook Air 15-inch · M5 · 10-core CPU / 10-core GPU · 16 GB unified
- MacBook Pro 14-inch (M5) · M5 · 10-core CPU / 10-core GPU · 16 GB unified
Less memory · 8 GB 3
More memory · 24 GB 8
- Mac mini M2 · 24 GB unified
- Mac mini M4 · 24 GB unified
- Mac mini M4 Pro · 24 GB unified
- Mac mini M5 Pro · 15-core CPU / 16-core GPU · 24 GB unified
- Mac mini M6 · 24 GB unified
- MacBook Air 13-inch M5 · 24 GB unified
- MacBook Pro 14-inch · M5 Pro · 18-core CPU / 20-core GPU · 24 GB unified
- MacBook Pro 16-inch · M5 Pro · 18-core CPU / 20-core GPU · 24 GB unified
