Best Local LLMs for 32GB RAM
Ranked open-weight models that run well on a 32GB machine, with estimated speed and the exact ollama command for each.
Try: Qwen3 8B, RTX 4090, MacBook Pro M4, 16GB, iPhone 17 Pro
Best local models for ~32GB
12 picksEstimates assume a representative Apple-Silicon machine with 32GB unified memory. Tok/s are ModelFit estimates, not measured benchmarks; typical error is the median gap to measured runs on hardware the model was not fitted on (how the estimates work). Run the wizard for figures tuned to your exact chip.
- 01
Gemma 4 26B-A4B
Best for chat, coding, multimodal. Strong fit for 32 GB RAM with balanced speed and quality.
$ollama run gemma4:26b- Fit
- Runs wellB · 29
- Best for
- Chat, Coding, Multimodal
- Perf
- est. 108 tok/s, typical error 10% · first token ~0.2s
- 02
Qwen3.8 27B
Best for coding, agent, vision, long context. Strong fit for 32 GB RAM with balanced speed and quality.
$ollama run qwen3.8:27b- Fit
- Runs wellB · 26
- Best for
- Coding, Agent, Vision, Long context
- Perf
- est. 30 tok/s, typical error 10% · first token ~1.3s
- 03
Nemotron 3.5 Lightning 30B-A3B (IQ4_XS)
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run nemotron-3.5-lightning-30b-q4xs- Fit
- Runs wellC · 16
- Best for
- Agentic, Coding, Long context
- Perf
- est. 132 tok/s, typical error 10% · first token ~0.2s
- 04
Qwen3.6 35B-A3B
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run qwen3.6:35b-a3b- Fit
- Runs wellC · 2
- Best for
- Reasoning, Coding, Agents
- Perf
- est. 163 tok/s, typical error 10% · first token ~0.2s
- 05
LFM2 24B-A2B Instruct
Best for local ai agents, privacy-first tool calling, mcp workflows. Strong fit for 32 GB RAM with balanced speed and quality.
$ollama run lfm2:24b-a2b- Fit
- Runs wellB · 37
- Best for
- Local AI agents, privacy-first tool calling, MCP workflows
- Perf
- est. 178 tok/s, typical error 10% · first token ~0.1s
- 06
GLM-4.7-Flash 30B-A3B
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run glm-4.7-flash- Fit
- Runs wellC · 15
- Best for
- Coding, Agentic tasks, Reasoning
- Perf
- est. 116 tok/s, typical error 10% · first token ~0.2s
- 07
Gemma 4 31B
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run gemma4:31b- Fit
- Runs wellC · 11
- Best for
- Quality, Coding, Multimodal
- Perf
- est. 24 tok/s, typical error 10% · first token ~1.6s
- 08
Gemma 4 12B (Q8)
Best for chat, coding, multimodal. Strong fit for 32 GB RAM with balanced speed and quality.
$ollama run gemma4:12b-it-q8_0- Fit
- Runs wellB · 43
- Best for
- Chat, Coding, Multimodal
- Perf
- est. 37 tok/s, typical error 10% · first token ~0.7s
- 09
Laguna XS 2.1
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run laguna-xs-2.1:q4_K_M- Fit
- Runs wellC · 9
- Best for
- Agentic coding, Long-horizon tasks
- Perf
- est. 153 tok/s, typical error 10% · first token ~0.2s
- 10
Qwen3.6 27B
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run qwen3.6:27b- Fit
- Runs wellC · 20
- Best for
- Coding, Quality, Long context
- Perf
- est. 30 tok/s, typical error 10% · first token ~1.3s
- 11
Ornith 1.0 35B
This model may feel memory-heavy on 32 GB RAM, but it is still listed for balanced speed and quality.
$ollama run ornith:35b- Fit
- Runs wellC · 5
- Best for
- Agentic coding
- Perf
- est. 166 tok/s, typical error 10% · first token ~0.2s
- 12
Gemma 4 12B
Best for chat, coding, multimodal. Strong fit for 32 GB RAM with balanced speed and quality.
$ollama run gemma4:12b- Fit
- Lots of roomA · 64
- Best for
- Chat, Coding, Multimodal
- Perf
- est. 59 tok/s, typical error 10% · first token ~0.6s
Get told when a better model fits your MacBook Air M5 16 GB
Get one email when a new open-weight model is a better fit for this machine.
For: MacBook Air M5 16 GB
You'll also get the weekly ModelFit email. By signing up you agree to our Privacy Policy. Unsubscribe from either email anytime, in one click.
Browse by RAM tier
Want an exact recommendation?
The wizard tunes picks and speed estimates to your exact device, chip, and RAM.