Back to leaderboard

deepseek-r1:70b

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

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

OLLAMA GGUF

qwen3next · 79.7B · Q4_K_M

Good

Aug 20, 2026 · Intel Xeon® E5-2637 v3

Global Score
47 vs 78
Hardware Fit
51 vs 49
Quality Score
45 vs 91

Hardware

deepseek-r1:70b qwen3-next:80b-a3b…
MachineNVIDIA NVIDIA_DGX_SparkDell Inc. Precision Tower 7910
CPUCortex-X925Intel Xeon® E5-2637 v3
Cores2016
RAM122 GB96 GB
GPUDevice 2e12NVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060
OSUbuntu 24.04.4 LTSMicrosoft Windows 10 Pro 10.0.19045
Archarm64x64
Power Modeperformancebalanced

Performance

deepseek-r1:70b qwen3-next:80b-a3b…
Tokens/sec4.78.7
First chunk557 ms1524 ms
TTFT557 ms1.5 s
Load time52.6 s63.9 s
Memory usage49.5 GB48.3 GB
Memory %41%50%

HW Fit Score Breakdown

deepseek-r1:70b

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

qwen3-next:80b-a3b-inst…

Speed
9/50
TTFT
15/20
Memory
25/30

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

qwen3-next:80b-a3b-inst…

Reasoning
18/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
13/15
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