Back to leaderboard

devstral-small-2-24b-instruct-2512

LM-STUDIO GGUF

mistral3 · 24B · IQ4_XS

Excellent

Sep 20, 2026 · AMD Ryzen 7 5800X 8-Core Processor

qwen3-30b-a3b-thinking-2507-claude-4.5-sonnet-high-reasoning-distill-mlx

LM-STUDIO MLX

qwen3_moe · 30B · MXFP4

Excellent

Aug 20, 2026 · Apple M4 Max

Global Score
81 vs 96
Hardware Fit
80 vs 95
Quality Score
81 vs 96

Hardware

devstral-small-2-2… qwen3-30b-a3b-thin…
MachineSystem manufacturerMacBook Pro
CPUAMD Ryzen 7 5800X 8-Core ProcessorApple M4 Max
Cores1616
RAM31 GB128 GB
GPUNavi 44 [Radeon RX 9060 XT]Apple M4 Max
OSUbuntu 26.04.1 LTSmacOS 26.5.2
Archx64arm64
Power Modeperformancebalanced

Performance

devstral-small-2-2… qwen3-30b-a3b-thin…
Tokens/sec21.5112.8
First chunk12 ms21 ms
TTFT212 ms1.5 s
Load timeN/A3.5 s
Memory usage15.9 GB0.0 GB
Memory %51%0%

HW Fit Score Breakdown

devstral-small-2-24b-in…

Speed
35/50
TTFT
20/20
Memory
25/30

qwen3-30b-a3b-thinking-…

Speed
50/50
TTFT
15/20
Memory
30/30

Quality

devstral-small-2-24b-in…

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

qwen3-30b-a3b-thinking-…

Reasoning
19/20
Coding
18/20
Instruction
19/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong 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