Back to leaderboard

phi3:3.8b

OLLAMA GGUF

phi3 · 3.8B · Q4_0

Good

Mar 2, 2026 · Apple M4

codestral:22b

OLLAMA GGUF

llama · 22.2B · Q4_0

Good

Jun 26, 2026 · Cortex-X925

Global Score
64 vs 67
Hardware Fit
100 vs 67
Quality Score
48 vs 67

Hardware

phi3:3.8b codestral:22b
MachineMacBook AirNVIDIA NVIDIA_DGX_Spark
CPUApple M4Cortex-X925
Cores1020
RAM32 GB122 GB
GPUApple M4Device 2e12
OSmacOS 26.3Ubuntu 24.04.4 LTS
Archarm64arm64
Power Modebalancedperformance

Performance

phi3:3.8b codestral:22b
Tokens/sec42.017.2
First chunkN/A129 ms
TTFT142 ms129 ms
Load time1.1 s11.6 s
Memory usage16.0 GB18.9 GB
Memory %50%16%

HW Fit Score Breakdown

phi3:3.8b

Speed
40/50
TTFT
30/20
Memory
30/30

codestral:22b

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

Quality

phi3:3.8b

Reasoning
12/20
Coding
3/20
Instruction
11/20
Structured
10/15
Math
5/15
Multilingual
7/10
Reasoning: Adequate Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Weak Multilingual: Adequate

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

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