Back to leaderboard

deepseek-r1:70b

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

gemma4:12b

OLLAMA GGUF

gemma4 · 11.9B · Q4_K_M

Excellent

Jul 4, 2026 · Apple M2 Pro

Global Score
47 vs 89
Hardware Fit
51 vs 88
Quality Score
45 vs 89

Hardware

deepseek-r1:70b gemma4:12b
MachineNVIDIA NVIDIA_DGX_SparkMacBook Pro
CPUCortex-X925Apple M2 Pro
Cores2010
RAM122 GB16 GB
GPUDevice 2e12Apple M2 Pro
OSUbuntu 24.04.4 LTSmacOS 26.5
Archarm64arm64
Power Modeperformancebalanced

Performance

deepseek-r1:70b gemma4:12b
Tokens/sec4.718.4
First chunk557 ms758 ms
TTFT557 ms758 ms
Load time52.6 s0.3 s
Memory usage49.5 GB7.5 GB
Memory %41%47%

HW Fit Score Breakdown

deepseek-r1:70b

Speed
4/50
TTFT
20/20
Memory
27/30

gemma4:12b

Speed
42/50
TTFT
20/20
Memory
26/30

Quality

deepseek-r1:70b

Reasoning
3/20
Coding
1/20
Instruction
14/20
Structured
9/15
Math
8/15
Multilingual
10/10
Reasoning: Poor Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Adequate Multilingual: Strong

gemma4:12b

Reasoning
17/20
Coding
18/20
Instruction
17/20
Structured
15/15
Math
12/15
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