PipelineScore
← Back to user leaderboard
User profile

gpu-shopper

12 runsFirst seen 2026-04-10Avg 64.3
Best PipelineScore
85.1MAINLINE
Total tokens67.4KAcross every task this user has run
Avg latency758msPer task, across all submissions
Tasks run30012 submissions x ~25 tasks
Rigs used10Distinct hardware tags

Category signature

Average score per category across all 12 runs.

Code
64.5
Reason
62.6
Write
65.8
Tool Use
65.1
RAG
63.9
Speed
64.5

Hardware mix

Rigs this user benchmarked on.

macbook-pro-m3-pro-18gb3 (25%)
h200-141gb1 (8%)
b200-192gb1 (8%)
m3-max-64gb1 (8%)
m3-pro-36gb1 (8%)
ryzen-7950x-cpu-only1 (8%)
a100-40gb1 (8%)
m1-air-8gb1 (8%)
rtx-4070-12gb1 (8%)
rtx-3060-12gb1 (8%)

Provider mix

Where they spend their tokens.

alibaba2 (17%)
microsoft2 (17%)
cohere2 (17%)
cognitivecomputations1 (8%)
zyphra1 (8%)
upstage1 (8%)
google1 (8%)
community1 (8%)
bigcode1 (8%)

Models tried

Best score per model. Click a model to see its full page.

#ModelBest ScoreTier
1Qwen 2.5 72B Instruct85.1MAINLINE
2WizardLM 2 8x22B79.5MAINLINE
3Aya 23 35B74.1FEEDER
4Dolphin 3.0 Llama 3.1 8B68.5FEEDER
5Phi 4 14B67.9FEEDER
6Zamba 2 7B Instruct66.7FEEDER
7SOLAR 10.7B Instruct66.6FEEDER
8Command R65.5FEEDER
9Gemma 3 4B IT58.8TAP
10LLaVA 1.6 Mistral 7B56.0TAP
11StarCoder2 3B49.5TAP
12Qwen 2.5 0.5B Instruct33.9DRIP

All submissions

Every run, ordered by score.

#ModelScoreTier
1Qwen 2.5 72B Instruct85.1MAINLINE
2WizardLM 2 8x22B79.5MAINLINE
3Aya 23 35B74.1FEEDER
4Dolphin 3.0 Llama 3.1 8B68.5FEEDER
5Phi 4 14B67.9FEEDER
6Zamba 2 7B Instruct66.7FEEDER
7SOLAR 10.7B Instruct66.6FEEDER
8Command R65.5FEEDER
9Gemma 3 4B IT58.8TAP
10LLaVA 1.6 Mistral 7B56.0TAP
11StarCoder2 3B49.5TAP
12Qwen 2.5 0.5B Instruct33.9DRIP
gpu-shopper · PipelineScore