Best local LLMs for the Apple M5 Ultra (512 GB)

Each list assumes the context that task needs, so a model that fits for chat may rank lower for long documents.

General chat at 8K context

  1. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q4_K_M · 57.0 tok/s · Runs great
  2. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Intern-S2 (397B-A17B) Q6_K · 43.6 tok/s · Runs well (tight fit)
  3. 91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. K2-Horizon (375B-A23B) Q5_K_M · 39.3 tok/s · Runs great
  4. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Ling 3.1 Flash 1019 (560B-A25B) Q4_K_M · 43.3 tok/s · Runs well (tight fit)
  5. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2 Flash (309B-A15B) Q6_K · 48.4 tok/s · Runs great

Coding at 32K context

  1. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q4_K_M · 57.0 tok/s · Runs great
  2. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Intern-S2 (397B-A17B) Q6_K · 43.6 tok/s · Runs well (tight fit)
  3. 91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. K2-Horizon (375B-A23B) Q5_K_M · 39.3 tok/s · Runs great
  4. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Ling 3.1 Flash 1019 (560B-A25B) Q4_K_M · 43.3 tok/s · Runs well (tight fit)
  5. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2 Flash (309B-A15B) Q6_K · 48.4 tok/s · Runs great

Reasoning at 32K context

  1. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q4_K_M · 57.0 tok/s · Runs great
  2. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Intern-S2 (397B-A17B) Q6_K · 43.6 tok/s · Runs well (tight fit)
  3. 91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. K2-Horizon (375B-A23B) Q5_K_M · 39.3 tok/s · Runs great
  4. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Ling 3.1 Flash 1019 (560B-A25B) Q4_K_M · 43.3 tok/s · Runs well (tight fit)
  5. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2 Flash (309B-A15B) Q6_K · 48.4 tok/s · Runs great

Documents (RAG) at 32K context

  1. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q4_K_M · 57.0 tok/s · Runs great
  2. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Intern-S2 (397B-A17B) Q6_K · 43.6 tok/s · Runs well (tight fit)
  3. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2 Flash (309B-A15B) Q6_K · 48.4 tok/s · Runs great
  4. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiniMax M2 (230B-A10B) Q8_0 · 51.8 tok/s · Runs great
  5. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiniMax M2.7 (230B-A10B) Q6_K · 66.8 tok/s · Runs great

Agents & tools at 64K context

  1. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q4_K_M · 57.0 tok/s · Runs great
  2. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Intern-S2 (397B-A17B) Q6_K · 43.6 tok/s · Runs well (tight fit)
  3. 91Excellent quality (A+) · 91 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. K2-Horizon (375B-A23B) Q5_K_M · 39.3 tok/s · Runs great
  4. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. Ling 3.1 Flash 1019 (560B-A25B) Q4_K_M · 43.3 tok/s · Runs well (tight fit)
  5. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 100/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2 Flash (309B-A15B) Q6_K · 48.4 tok/s · Runs great

Reverse engineering at 32K context

  1. 96Excellent quality (A+) · 96 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. Nex N2.6 (321B-A18B) Q6_K · 41.6 tok/s · Runs great
  2. 95Excellent quality (A+) · 95 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. GLM-5.3 Flash (320B-A18B) Q6_K · 41.6 tok/s · Runs great
  3. 95Excellent quality (A+) · 95 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 98/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2.6 Flash (310B-A15B) Q6_K · 48.4 tok/s · Runs great
  4. 84Great quality (A) · 84 of 100: how capable the model is at F16, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. gpt-oss 120B F16 · 46.5 tok/s · Runs great
  5. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking. Naive N0.5 Flash (309B-A15.5B) Q6_K · 47.1 tok/s · Runs great

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?