Best local LLMs for the Marvell OCTEON TX2 CN9130 NAS (32 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. 94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 27B Q6_K · 0.2 tok/s · Very slow (tight fit)
  2. 93Excellent quality (A+) · 93 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q6_K · 0.2 tok/s · Very slow (tight fit)
  3. 92Excellent quality (A+) · 92 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 27B Q6_K · 0.2 tok/s · Very slow (tight fit)
  4. 90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Nanbeige4.3 33B Q4_K_M · 0.2 tok/s · Very slow (tight fit)
  5. 87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Muse Glimmer 30B Q4_K_M · 0.3 tok/s · Very slow

Coding at 32K context

  1. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q4_K_M · 0.3 tok/s · Very slow
  2. 87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Muse Glimmer 30B Q4_K_M · 0.3 tok/s · Very slow
  3. 78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM-V 4.7 35B-A3B Q3_K_M · 3.5 tok/s · Slow
  4. 77Great quality (A) · 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. Nex N2 Mini (35B-A3B) Q3_K_M · 3.5 tok/s · Slow
  5. 76Great quality (A) · 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. KAT-Coder V2.5 Dev (35B-A3B) Q3_K_M · 3.5 tok/s · Slow

Reasoning at 32K context

  1. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q4_K_M · 0.3 tok/s · Very slow
  2. 87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Muse Glimmer 30B Q4_K_M · 0.3 tok/s · Very slow
  3. 78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM-V 4.7 35B-A3B Q3_K_M · 3.5 tok/s · Slow
  4. 77Great quality (A) · 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. Nex N2 Mini (35B-A3B) Q3_K_M · 3.5 tok/s · Slow
  5. 83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. Fara 1.5 27B Q4_K_M · 0.3 tok/s · Very slow

Documents (RAG) at 32K context

  1. 90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 27B Q4_K_M · 0.3 tok/s · Very slow
  2. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q4_K_M · 0.3 tok/s · Very slow
  3. 88Excellent quality (A+) · 88 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 27B Q4_K_M · 0.3 tok/s · Very slow
  4. 87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Muse Glimmer 30B Q4_K_M · 0.3 tok/s · Very slow
  5. 85Excellent quality (A+) · 85 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. TMax 27B Q4_K_M · 0.3 tok/s · Very slow

Agents & tools at 64K context

  1. 87Excellent quality (A+) · 87 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Muse Glimmer 30B Q4_K_M · 0.3 tok/s · Very slow
  2. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 85/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q4_K_M · 0.3 tok/s · Very slow (tight fit)
  3. 78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM-V 4.7 35B-A3B Q3_K_M · 3.5 tok/s · Slow
  4. 77Great quality (A) · 77 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. Nex N2 Mini (35B-A3B) Q3_K_M · 3.5 tok/s · Slow
  5. 76Great quality (A) · 76 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. KAT-Coder V2.5 Dev (35B-A3B) Q3_K_M · 3.5 tok/s · Slow

Reverse engineering at 32K context

  1. 88Excellent quality (A+) · 88 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 86/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 27B Q4_K_M · 0.3 tok/s · Very slow
  2. 78Great quality (A) · 78 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.5 35B-A3B Q3_K_M · 3.5 tok/s · Slow
  3. 73Great quality (A) · 73 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. GLM-4.7 Flash (30B-A3B) Q3_K_M · 3.5 tok/s · Slow
  4. 90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.8 27B Q4_K_M · 0.3 tok/s · Very slow
  5. 89Excellent quality (A+) · 89 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. Qwen3.6 27B Q4_K_M · 0.3 tok/s · Very slow

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?