Back to leaderboard

qwen3.6-35b-a3b-claude-4.7-opus-reasoning-distilled-apex-mtp

LM-STUDIO GGUF

qwen35moe · 35B

Excellent

Jun 19, 2026 · AMD Ryzen 5 5600X 6-Core Processor

mistralai/devstral-small-2-2512

LM-STUDIO MLX

mistral3 · 24B · 4bit

Good

Oct 20, 2026 · Apple M1 Pro

Global Score
92 vs 73
Hardware Fit
87 vs 64
Quality Score
94 vs 77

Hardware

qwen3.6-35b-a3b-cl… mistralai/devstral…
MachineGigabyte Technology Co., Ltd. B550M AORUS PROMacBook Pro
CPUAMD Ryzen 5 5600X 6-Core ProcessorApple M1 Pro
Cores1210
RAM47 GB32 GB
GPUAD106 [GeForce RTX 4060 Ti]Apple M1 Pro
OScachyos rollingmacOS 15.6.1
Archx64arm64
Power Modelow-powerbalanced

Performance

qwen3.6-35b-a3b-cl… mistralai/devstral…
Tokens/sec42.611.9
First chunk13 ms7 ms
TTFT3.1 s1.1 s
Load timeN/AN/A
Memory usage23.2 GB19.7 GB
Memory %49%62%

HW Fit Score Breakdown

qwen3.6-35b-a3b-claude-…

Speed
50/50
TTFT
12/20
Memory
25/30

mistralai/devstral-smal…

Speed
25/50
TTFT
20/20
Memory
19/30

Quality

qwen3.6-35b-a3b-claude-…

Reasoning
20/20
Coding
16/20
Instruction
19/20
Structured
15/15
Math
14/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

mistralai/devstral-smal…

Reasoning
17/20
Coding
16/20
Instruction
13/20
Structured
14/15
Math
7/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest