Back to leaderboard

codestral:22b

OLLAMA GGUF

llama · 22.2B · Q4_0

Good

Jun 26, 2026 · Cortex-X925

qwen3.5-27b-claude-4.6-opus-distilled-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 4bit

Not Rec.

Mar 10, 2026 · Apple M4

Global Score
67 vs 34
Hardware Fit
67 vs 21
Quality Score
67 vs 39

Hardware

codestral:22b qwen3.5-27b-claude…
MachineNVIDIA NVIDIA_DGX_SparkMac mini
CPUCortex-X925Apple M4
Cores2010
RAM122 GB24 GB
GPUDevice 2e12Apple M4
OSUbuntu 24.04.4 LTSmacOS 26.3
Archarm64arm64
Power Modeperformancebalanced

Performance

codestral:22b qwen3.5-27b-claude…
Tokens/sec17.26.1
First chunk129 ms7 ms
TTFT129 ms22.9 s
Load time11.6 sN/A
Memory usage18.9 GB19.8 GB
Memory %16%82%

HW Fit Score Breakdown

codestral:22b

Speed
17/50
TTFT
20/20
Memory
30/30

qwen3.5-27b-claude-4.6-…

Speed
12/50
TTFT
0/20
Memory
9/30

Quality

codestral:22b

Reasoning
11/20
Coding
17/20
Instruction
9/20
Structured
14/15
Math
6/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak Structured Output: Strong Math: Weak Multilingual: Strong

qwen3.5-27b-claude-4.6-…

Reasoning
1/20
Coding
14/20
Instruction
5/20
Structured
8/15
Math
4/15
Multilingual
7/10
Reasoning: Poor Coding: Adequate Instruction Following: Weak Structured Output: Adequate Math: Poor Multilingual: Adequate

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