User profile

drew

Beta #1

5 runsFirst seen 2026-07-06Avg 87.5

Best PipelineScore

93.2TRUNK

Best run beats 97% of all 34 submissionsBest rig m5-max-48gb ranks #1 of 10 rigs

Total tokens73.4KAcross every task this user has run

Avg latency9282msPer task, across all submissions

Tasks run1705 submissions x ~34 tasks

Rigs used2Distinct hardware tags

Category signature

Average score per category across all 5 runs.

Code

95.0

Reason

82.0

Tool Use

87.5

RAG

100.0

Speed

47.6

Rigs this user benchmarked on.

m5-max-48gb3 (60%)

m3-ultra-96gb2 (40%)

Where they spend their tokens.

local5 (100%)

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

#	Model	Provider	Best Score	Tier	Achieved
1	openai/gpt-oss-20b	local	93.2	TRUNK	2026-07-06
2	muse-glimmer:30b	local	84.8	MAINLINE	2026-10-01
3	mlx-community/Qwen3.6-35B-A3B-4bit	local	83.7	MAINLINE	2026-07-06

Every run, ordered by score.

#	Model	Hardware	Score	Tier	Code	Reason	Tool Use	RAG	Speed	Tokens	Avg ms	Date
1	openai/gpt-oss-20b	m5-max-48gb	93.2	TRUNK	100.0	80.0	87.5	100.0	99.3	4.2K	1097	2026-07-06
2	openai/gpt-oss-20b	m5-max-48gb	90.8	TRUNK	100.0	70.0	87.5	100.0	0.0	3.7K	1066	2026-07-06
3	muse-glimmer:30b	m3-ultra-96gb	84.8	MAINLINE	100.0	90.0	87.5	100.0	28.7	14.9K	15230	2026-10-01
4	muse-glimmer:30b	m5-max-48gb	84.7	MAINLINE	100.0	90.0	87.5	100.0	27.8	14.9K	15552	2026-10-01
5	mlx-community/Qwen3.6-35B-A3B-4bit	m3-ultra-96gb	83.7	MAINLINE	75.0	80.0	87.5	100.0	82.5	35.7K	13463	2026-07-06