Back to leaderboard

qwen3-next:80b-a3b-instruct-q4_K_M

OLLAMA GGUF

qwen3next · 79.7B · Q4_K_M

Not Rec.

Aug 23, 2026 · Intel Xeon® E5-4650 v2

lfm2.5-1.2b-instruct-mlx

LM-STUDIO MLX

lfm2 · 1.2B · 8bit

Good

Mar 4, 2026 · Apple M4

Global Score
74 vs 68
Hardware Fit
42 vs 100
Quality Score
87 vs 54

Hardware

qwen3-next:80b-a3b… lfm2.5-1.2b-instru…
MachineDell Inc. PowerEdge R820MacBook Air
CPUIntel Xeon® E5-4650 v2Apple M4
Cores8010
RAM256 GB32 GB
GPUMicrosoft Basic Display AdapterApple M4
OSMicrosoft Windows Server 2022 Datacenter 10.0.20348macOS 26.3
Archx64arm64
Power Modebalancedbalanced

Performance

qwen3-next:80b-a3b… lfm2.5-1.2b-instru…
Tokens/sec3.977.3
First chunk2089 ms177 ms
TTFT2.1 s177 ms
Load time73.3 sN/A
Memory usage46.9 GB1.2 GB
Memory %18%4%

HW Fit Score Breakdown

qwen3-next:80b-a3b-inst…

Speed
3/50
TTFT
9/20
Memory
30/30

lfm2.5-1.2b-instruct-mlx

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

Quality

qwen3-next:80b-a3b-inst…

Reasoning
18/20
Coding
19/20
Instruction
12/20
Structured
15/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong

lfm2.5-1.2b-instruct-mlx

Reasoning
10/20
Coding
7/20
Instruction
16/20
Structured
12/15
Math
1/15
Multilingual
8/10
Reasoning: Weak Coding: Weak Instruction Following: Strong Structured Output: Strong Math: Poor Multilingual: Strong
View qwen3-next:80b-a3b… details | View lfm2.5-1.2b-instru… 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