Back to leaderboard

google/gemma-3-12b

gemma3 · 12B · 4bit

Good

Mar 3, 2026 · Apple M4

mistralai/magistral-small-2509

mistral3 · 24B · 4bit

Not Rec.

Mar 3, 2026 · Apple M4

Global Score
77 vs 20
Hardware Fit
72 vs 65
Quality Score
79 vs 0

Hardware

google/gemma-3-12b mistralai/magistra…
MachineMacBook AirMacBook Air
CPUApple M4Apple M4
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.3macOS 26.3
Archarm64arm64
Power Modebalancedbalanced

Performance

google/gemma-3-12b mistralai/magistra…
Tokens/sec7.95.1
TTFT1.8 s2.0 s
Load time0.0 s0.0 s
Memory usage0.0 GB0.0 GB
Memory %0%0%

HW Fit Score Breakdown

google/gemma-3-12b

Speed
14/40
TTFT
28/30
Memory
30/30

mistralai/magistral-sma…

Speed
8/40
TTFT
27/30
Memory
30/30

Quality

google/gemma-3-12b

Reasoning
13/20
Coding
19/20
Instruction
14/20
Structured
15/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Adequate Multilingual: Strong

mistralai/magistral-sma…

Reasoning
0/20
Coding
0/20
Instruction
0/20
Structured
0/15
Math
0/15
Multilingual
0/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running