Best Local LLMs for 36GB RAM

Ranked open-weight models that run well on a 36GB machine, with estimated speed and the exact ollama command for each.

Try: RTX 4090, MacBook Pro M4, 16GB, iPhone 17 Pro

Best local models for ~36GB

12 picks

Estimates assume a representative Apple-Silicon machine with 36GB unified memory. Tok/s are ModelFit estimates, not measured benchmarks. Run the wizard for figures tuned to your exact chip.

01

Qwen3.5 35B-A3B Instruct

Qwen / 35B / Q4_K_M / ~20 GB

Best for: Reasoning, Coding, Agent scenarios·Perf: ~128.3 tok/s · first token ~1.3s

Runs well

This model may feel memory-heavy on 36 GB RAM, but it is still listed for balanced speed and quality.

ollama
$ollama run qwen3.5:35b-a3b
02

Gemma 4 26B-A4B

Gemma / 26B / Q4_K_M / ~16 GB

Best for: Chat, Coding, Multimodal·Perf: ~129.5 tok/s · first token ~0.5s

Runs well

Best for chat, coding, multimodal. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run gemma4:26b
03

Qwen3.5 27B Instruct

Qwen / 27B / Q4_K_M / ~16 GB

Best for: Chat, Coding, Complex reasoning·Perf: ~53.9 tok/s · first token ~0.6s

Runs well

Best for chat, coding, complex reasoning. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run qwen3.5:27b
04

Qwen3.6 27B

Qwen / 27B / Q4_K_M / ~18 GB

Best for: Coding, Quality, Long context·Perf: ~53.9 tok/s · first token ~0.6s

Runs well

Best for coding, quality, long context. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run qwen3.6:27b
05

GPT-OSS 20B

GPT-OSS / 21B / MXFP4 / ~13.8 GB

Best for: Chat, Coding, Reasoning·Perf: ~134.5 tok/s · first token ~0.5s

Runs well

Best for chat, coding, reasoning. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run gpt-oss:20b
06

LFM2 24B-A2B Instruct

LFM2 / 24B / Q4_K_M / ~14 GB

Best for: Local AI agents, privacy-first tool calling, MCP workflows·Perf: ~180 tok/s · first token ~0.5s

Runs well

Best for local ai agents, privacy-first tool calling, mcp workflows. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run lfm2:24b-a2b
07

Qwen3.6 35B-A3B

Qwen / 35B / Q4_K_M / ~22 GB

Best for: Reasoning, Coding, Agents·Perf: ~116.6 tok/s · first token ~1.3s

Runs well

This model may feel memory-heavy on 36 GB RAM, but it is still listed for balanced speed and quality.

ollama
$ollama run qwen3.6:35b-a3b
08

Gemma 4 31B

Gemma / 31B / Q4_K_M / ~20 GB

Best for: Quality, Coding, Multimodal·Perf: ~47.4 tok/s · first token ~1.5s

Runs well

This model may feel memory-heavy on 36 GB RAM, but it is still listed for balanced speed and quality.

ollama
$ollama run gemma4:31b
09

Gemma 4 12B (Q8)

Gemma / 12B / Q8_0 / ~12.8 GB

Best for: Chat, Coding, Multimodal·Perf: ~69.4 tok/s · first token ~0.6s

Runs well

Best for chat, coding, multimodal. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run gemma4:12b-it-q8_0
10

Gemma 3 27B Instruct

Gemma / 27B / Q4_K_M / ~21 GB

Best for: Quality, Coding·Perf: ~51.1 tok/s · first token ~0.6s

Runs well

This model may feel memory-heavy on 36 GB RAM, but it is still listed for balanced speed and quality.

ollama
$ollama run gemma3:27b
11

Qwen3 30B

Qwen / 30B / Q4_K_M / ~22 GB

Best for: Quality, Coding·Perf: ~125 tok/s · first token ~1.3s

Runs well

This model may feel memory-heavy on 36 GB RAM, but it is still listed for balanced speed and quality.

ollama
$ollama run qwen3:30b
12

Mistral Small 3.1

Mistral / 24B / Q4_K_M / ~15 GB

Best for: Chat, Coding·Perf: ~59.9 tok/s · first token ~0.6s

Runs well

Best for chat, coding. Strong fit for 36 GB RAM with balanced speed and quality.

ollama
$ollama run mistral-small3.1:24b

Browse by RAM tier

8 GB16 GB24 GB32 GB48 GB64 GB96 GB128 GB

Want an exact recommendation?

The wizard tunes picks and speed estimates to your exact device, chip, and RAM.

Open ModelFit Wizard Browse all devices