Back to leaderboard

gemma-4-26b-a4b-it-qat

LM-STUDIO GGUF

gemma4 · 26B · Q4_K_XL

Excellent

Jul 7, 2026 · AMD Ryzen 7 7800X3D 8-Core Processor

deepseek-r1:70b

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Jun 27, 2026 · Cortex-X925

Global Score
88 vs 47
Hardware Fit
77 vs 51
Quality Score
93 vs 45

Hardware

gemma-4-26b-a4b-it… deepseek-r1:70b
MachineASUSNVIDIA NVIDIA_DGX_Spark
CPUAMD Ryzen 7 7800X3D 8-Core ProcessorCortex-X925
Cores1620
RAM31 GB122 GB
GPUNVIDIA GeForce RTX 4090, RaphaelDevice 2e12
OScachyos rollingUbuntu 24.04.4 LTS
Archx64arm64
Power Modeperformanceperformance

Performance

gemma-4-26b-a4b-it… deepseek-r1:70b
Tokens/sec169.64.7
First chunk9 ms557 ms
TTFT108 ms557 ms
Load timeN/A52.6 s
Memory usage26.0 GB49.5 GB
Memory %85%41%

HW Fit Score Breakdown

gemma-4-26b-a4b-it-qat

Speed
50/50
TTFT
20/20
Memory
7/30

deepseek-r1:70b

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

Quality

gemma-4-26b-a4b-it-qat

Reasoning
19/20
Coding
19/20
Instruction
18/20
Structured
14/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

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
View gemma-4-26b-a4b-it… details | View deepseek-r1:70b 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