[mle]/DaBench

Apodex 1.1 Mini (free) vs Muse Spark 1.3 Contributor

Side A
Model page
Side B
Model page
Overall
25
vs
Overall
66

Muse Spark 1.3 Contributor is clearly better. It wins 75 of the 77 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 1.3× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
2535.1s
8.5K, 75% thinking8.7K, 91% thinking
661m 42s
Agents
2639.4s
6.2K, 69% thinking14.7K, 88% thinking
573m 16s
Structured Extraction
5227.2s
7.7K, 68% thinking4.2K, 91% thinking
9037.9s
Coding
446.5s
12.1K, 94% thinking10.8K, 95% thinking
482m 1s
Grounded Answers
2926.0s
6.9K, 70% thinking6.1K, 91% thinking
811m 1s
Instruction Following
3644.5s
11.4K, 79% thinking8.7K, 91% thinking
841m 18s
Business Logic
1435.4s
10K, 56% thinking6K, 95% thinking
761m 7s
Security & permissions
4027.3s
5.8K, 76% thinking8.2K, 87% thinking
761m 59s
Diagnosis & investigation
034.6s
5.1K, 84% 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
278 tok/s (better)
107 tok/s
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
256K
1M (better)
Cost per run
Not measured
$0.41
Cost per case
Not measured
$0.0027
Time per case
35.1s (better)
1m 42s
Quantization
BF16
—
Harness
openrouter
openrouter

The race

2 lead changes

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

Apodex 1.1 Mini (free)Muse Spark 1.3 Contributor
025507510050100150Perfect run2566

Every test case

153 test cases

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

2 won by Apodex 1.1 Mini (free)75 won by Muse Spark 1.3 Contributor76 tied