Back to leaderboard

phi4-mini

OLLAMA GGUF
Not Rec.

Aug 28, 2026 · Intel Core™ i5-10210U

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

LM-STUDIO MLX

qwen3_moe · 30B · MXFP4

Excellent

Apr 22, 2026 · Apple M2 Max

Global Score
51 vs 95
Hardware Fit
46 vs 94
Quality Score
53 vs 96

Hardware

phi4-mini qwen3-30b-a3b-thin…
MachineLENOVO 20U30029MBMac Studio
CPUIntel Core™ i5-10210UApple M2 Max
Cores812
RAM15 GB96 GB
GPUCometLake-U GT2 [UHD Graphics]Apple M2 Max
OSPop 24.04 LTSmacOS 26.1
Archx64arm64
Power Modeperformancebalanced

Performance

phi4-mini qwen3-30b-a3b-thin…
Tokens/sec2.680.4
First chunk5406 ms46 ms
TTFT5.4 s1.8 s
Load time27.4 sN/A
Memory usage4.6 GB21.2 GB
Memory %30%22%

HW Fit Score Breakdown

phi4-mini

Speed
6/50
TTFT
10/20
Memory
30/30

qwen3-30b-a3b-thinking-…

Speed
50/50
TTFT
14/20
Memory
30/30

Quality

phi4-mini

Reasoning
11/20
Coding
9/20
Instruction
6/20
Structured
12/15
Math
5/15
Multilingual
10/10
Reasoning: Adequate Coding: Weak Instruction Following: Weak 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