qwen3.6:35b-mlx

nvfp4

THINKING MODEL

MacBook Pro (Apple M5 Max)

48 GB · macOS 26.5.2

Tested on July 17, 2026 · Submitted by local_llama

Top 35% Compare

Global Score

80 /100

Not Rec.

Hardware Fit

77/100

Quality

81/100

Get this model

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

MLX versions & conversions

Hardware

Machine: MacBook Pro
CPU: Apple M5 Max
Cores: 18 threads (6 cores)
Frequency: 2.4 GHz
RAM: 48 GB LPDDR5
GPU: Apple M5 Max
OS: macOS 26.5.2
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 129.8
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.0
First chunk latency Delay before the first streamed chunk arrives from runtime: 77 ms
Time to first token How fast the model starts responding — lower is better: 30.0 s
Load time Time to load the model into memory before inference: 1.8 s
Memory usage RAM consumed during inference — percentage of total system memory: 19.7 GB (41%)
Total tokens Total number of tokens generated across all benchmark prompts: 1429
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~786

Score breakdown

Speed

50/50

Time to first token

0/20

Memory

27/30

Quality

Reasoning

19/20

Coding

9/20

Instruction following

18/20

Structured output

11/15

Math

14/15

Multilingual

10/10

Category levels

Reasoning: Strong Coding: Weak Instruction Following: Strong Structured Output: Adequate Math: Strong Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.30.10
Model format: GGUF
Hardware profile: HIGH-END
Result hash: 576749acc3588edeb3468704030c2aabb628c88f40e4cd9fe8cb0a43098ac0ec

Interpretation

Hardware fit: 77/100. Overall suitability: NOT RECOMMENDED (Global 80/100). Category profile: Reasoning: Strong, Coding: Weak, Instruction Following: Strong, Structured Output: Adequate, Math: Strong, Multilingual: Strong.

Disqualifiers

Time to first token too high: 30000ms (maximum: 16341ms for HIGH-END profile)

Bench Environment

Power: AC CPU load: avg 7% (peak 8%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest