[mle]/DaBench

Claude Haiku 5.5 vs Muse Spark 1.3 Contributor

Side A
Model page
Side B
Model page
Overall
70
vs
Overall
66

Too close to call. Claude Haiku 5.5 wins 19 of the 33 test cases where they differ.

Learn to build evals like this in the AI Engineering Academy

Task by task

A score turns orange only when that model wins clearly more of the task's test cases. Under each score, the average time per test case on the machine it ran on. Muse Spark 1.3 Contributor spends 2.6× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
7025.5s
3.9K, 78% thinking8.7K, 91% thinking
661m 42s
Agents
6139.5s
3.7K, 61% thinking14.7K, 88% thinking
573m 16s
Structured Extraction
9013.5s
2.6K, 83% thinking4.2K, 91% thinking
9037.9s
Coding
8630.6s
6.8K, 76% thinking10.8K, 95% thinking
482m 1s
Grounded Answers
7613.9s
2.2K, 78% thinking6.1K, 91% thinking
811m 1s
Instruction Following
9431.5s
4.3K, 82% thinking8.7K, 91% thinking
841m 18s
Business Logic
6726.0s
5.5K, 95% thinking6K, 95% thinking
761m 7s
Security & permissions
6723.4s
2.7K, 70% thinking8.2K, 87% thinking
761m 59s
Diagnosis & investigation
1725.5s
2.3K, 50% thinking17.2K, 90% thinking
173m 34s

What you'd run

The build behind each score. Orange marks the faster, smaller, cheaper or longer-context side.

SpecAB
Speed
225 tok/s (better)
107 tok/s
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
1M
1M
Cost per run
$0.42
$0.41 (better)
Cost per case
$0.0028
$0.0027 (better)
Time per case
25.5s (better)
1m 42s
Quantization
—
—
Harness
openrouter
openrouter

The race

4 lead changes

Each model's running score as the test cases go by. The dashed line is a perfect run.

Claude Haiku 5.5Muse Spark 1.3 Contributor
025507510050100150Perfect run7066

Every test case

153 test cases

Each square is one test case, colored by the model that scored higher.

19 won by Claude Haiku 5.514 won by Muse Spark 1.3 Contributor120 tied