User profile

context-window

4 runsFirst seen 2026-04-29Avg 75.8

Best PipelineScore

89.8DRIP

Total tokens25.7KAcross every task this user has run

Avg latency690msPer task, across all submissions

Tasks run1004 submissions x ~25 tasks

Rigs used4Distinct hardware tags

Category signature

Average score per category across all 4 runs.

Code

75.6

Reason

78.1

Write

76.2

Tool Use

75.2

RAG

75.3

Speed

78.5

Rigs this user benchmarked on.

rtx-4090-24gb1 (25%)

rtx-3090-24gb1 (25%)

a100-80gb1 (25%)

cloud-api1 (25%)

Where they spend their tokens.

Anthropic2 (50%)

DeepSeek1 (25%)

Cohere1 (25%)

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

#	Model	Provider	Best Score	Tier	Achieved
1	Claude Opus 4.7	Anthropic	89.8	DRIP	2026-05-13
2	DeepSeek V4	DeepSeek	84.1	MAINLINE	2026-05-20
3	Claude Haiku 4.5	Anthropic	73.8	FEEDER	2026-04-29
4	Command R+	Cohere	55.4	TAP	2026-05-06

Every run, ordered by score.

#	Model	Hardware	Score	Tier	Code	Reason	Write	Tool Use	RAG	Speed	Tokens	Avg ms	Date
1	Claude Opus 4.7	rtx-4090-24gb	89.8	DRIP	91.6	93.1	90.5	89.9	89.0	82.6	5.2K	1057	2026-05-13
2	DeepSeek V4	rtx-3090-24gb	84.1	MAINLINE	90.6	87.0	77.9	83.2	80.5	85.3	6.0K	412	2026-05-20
3	Claude Haiku 4.5	a100-80gb	73.8	FEEDER	70.5	75.1	75.9	73.3	72.4	92.8	6.8K	567	2026-04-29
4	Command R+	cloud-api	55.4	TAP	49.9	57.1	60.6	54.5	59.3	53.3	7.6K	722	2026-05-06