Skip to the comparison
STET
Get a free trial

DeepSeek V4.1 Flash vs Haiku 5.5 low+guidance

DeepSeek V4.1 Flash leads on quality: 5.7 points higher; Haiku 5.5 low+guidance leads on cost and agent time: 1.7× cheaper and 4.0× faster.

24 tasksData from

Stet score

DeepSeek V4.1 Flashdeepseek-harness · max reasoningRank 9 of 15

48.195% range 37.6–58.55.7 points higher

Haiku 5.5 low+guidanceclaude-code · low reasoningRank 10 of 15

42.495% range 33.2–51.7

A directional read: their 95% score ranges overlap.

Cost / task

DeepSeek V4.1 Flash$0.49from 23/24 tasks
Haiku 5.5 low+guidance$0.291.7× cheaper

Agent time / task

DeepSeek V4.1 Flash49.3 min
Haiku 5.5 low+guidance12.4 min4.0× faster

Quality

Head-to-head matchups

DeepSeek V4.1 Flash won9

Tied8

Haiku 5.5 low+guidance won7

DeepSeek V4.1 Flash’s win share 52%, 95% range 35–68%Ties are within judge noise.

Task by task

DeepSeek V4.1 Flash solved 3 tasks that Haiku 5.5 low+guidance didn’t, and Haiku 5.5 low+guidance solved 0 that DeepSeek V4.1 Flash didn’t. Both solved 9; neither solved 12.

DeepSeek V4.1 Flash only 3

  • stet-15439c21Passed tests
  • stet-commit-5d766ddePassed tests
  • stet-commit-342d0ac0Equivalent to the merged fix

Haiku 5.5 low+guidance only 0

A task counts as solved when its tests passed or a judge found the fix equivalent to the merged change, as in the leaderboard’s Solved column. Stet score also weighs code quality, so it can order the two differently.

All 24 compared tasks
Task by task
TaskDeepSeek V4.1 FlashHaiku 5.5 low+guidanceSolved by
stet-3169aefaPassed testsPassed testsBoth
stet-89dfbc27Passed testsPassed testsBoth
stet-b28f4fe7Passed testsPassed testsBoth
stet-232b50baPassed testsPassed testsBoth
stet-5276aa3fPassed testsPassed testsBoth
stet-commit-bbbbae09Passed testsPassed testsBoth
stet-66762b4cPassed testsPassed testsBoth
stet-15439c21Passed testsFailedDeepSeek V4.1 Flash only
stet-commit-f0caded9Passed testsEquivalent to the merged fixBoth
stet-commit-5d766ddePassed testsFailedDeepSeek V4.1 Flash only
stet-43493c52Passed testsPassed testsBoth
stet-6f84e978FailedFailedNeither
stet-commit-f8ae1177FailedFailedNeither
stet-2450ca2dFailedFailedNeither
stet-commit-342d0ac0Equivalent to the merged fixFailedDeepSeek V4.1 Flash only
stet-commit-60a0e6c2FailedFailedNeither
stet-commit-5e075641FailedFailedNeither
stet-commit-2826765eFailedFailedNeither
stet-commit-a8716b88FailedFailedNeither
stet-commit-30506f8eFailedFailedNeither
stet-commit-87814583FailedFailedNeither
stet-commit-d2ab0c25FailedFailedNeither
stet-commit-e41c2e2dFailedFailedNeither
stet-commit-e928166fFailedFailedNeither

Side by side

Side by side
MetricDeepSeek V4.1 FlashHaiku 5.5 low+guidance
Harnessdeepseek-harnessclaude-code
Reasoning effortmaxlow
Rank9 of 1510 of 15
Stet score48.195% range 37.6–58.542.495% range 33.2–51.7
Solved12/2411 passed, 1 judged equivalent9/248 passed, 1 judged equivalent
Cost / task$0.49from 23/24 tasks$0.29
Agent time / task49.3 min12.4 min

See which setup wins on your repo

Stet replays your merged PRs, scores quality above pass/fail, and delivers comparison reports through your agent.

Run it on your repo

About this data

  • 24 tasks, each a merged change replayed from the repo as it was just before the merge.
  • Data from .
  • Both setups come from the same leaderboard run of 15 setups: bmk_d4ba310fdf50db4ccb2092e3.

See all 15 setups on the leaderboardHow Stet measures