Back to leaderboard

granite-3.3-2b-instruct

LM-STUDIO MLX

granite · 2B · bf16

Good

Mar 4, 2026 · Apple M4

qwen3.8-9b-heretic-uncensored-nvfp4

LM-STUDIO GGUF

qwen35 · 9B · Q8_0

Excellent

Aug 20, 2026 · AMD Ryzen 7 9700X 8-Core Processor

Global Score
66 vs 82
Hardware Fit
87 vs 100
Quality Score
57 vs 74

Hardware

granite-3.3-2b-ins… qwen3.8-9b-heretic…
MachineMacBook AirMicro-Star International Co., Ltd. MS-7E51
CPUApple M4AMD Ryzen 7 9700X 8-Core Processor
Cores1016
RAM32 GB31 GB
GPUApple M4NVIDIA GeForce RTX 5080, AMD Radeon(TM) Graphics
OSmacOS 26.3Microsoft Windows 11 Pro 10.0.26200
Archarm64x64
Power Modebalancedbalanced

Performance

granite-3.3-2b-ins… qwen3.8-9b-heretic…
Tokens/sec18.3130.1
First chunk541 ms20 ms
TTFT541 ms99 ms
Load timeN/AN/A
Memory usage4.7 GB6.5 GB
Memory %15%21%

HW Fit Score Breakdown

granite-3.3-2b-instruct

Speed
37/50
TTFT
20/20
Memory
30/30

qwen3.8-9b-heretic-unce…

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

Quality

granite-3.3-2b-instruct

Reasoning
10/20
Coding
10/20
Instruction
14/20
Structured
14/15
Math
2/15
Multilingual
7/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Adequate

qwen3.8-9b-heretic-unce…

Reasoning
14/20
Coding
14/20
Instruction
13/20
Structured
14/15
Math
9/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Adequate Multilingual: Strong
View granite-3.3-2b-ins… details | View qwen3.8-9b-heretic… details

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