[mle]/DaBench

Apodex 1.1 Mini (free) vs Clover 35B A3B Preview

Side A
Model page
Side B
Model page
Overall
25
vs
Overall
30

Clover 35B A3B Preview is clearly better. It wins 21 of the 32 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. The time under each score is the average per test case, on the machine it ran on. Clover 35B A3B Preview spends 1.3× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
2535.1s
8.5K, 75% thinking8.4K, 95% thinking
3035.1s
Agents
2639.4s
6.2K, 69% thinking9.6K, 91% thinking
2036.8s
Structured Extraction
5227.2s
7.7K, 68% thinking7.2K, 97% thinking
4327.5s
Coding
446.5s
12.1K, 94% thinking12.2K, 100% thinking
101m 5s
Grounded Answers
2926.0s
6.9K, 70% thinking5.4K, 94% thinking
4821.0s
Instruction Following
3644.5s
11.4K, 79% thinking9.2K, 95% thinking
5332.8s
Business Logic
1435.4s
10K, 56% thinking9.8K, 99% thinking
1936.1s
Security & permissions
4027.3s
5.8K, 76% thinking5.8K, 92% thinking
4725.6s
Diagnosis & investigation
034.6s
5.1K, 84% thinking7K, 82% thinking
037.0s

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)
252 tok/s
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
256K
—
Cost per run
Not measured
Not measured
Cost per case
Not measured
Not measured
Time per case
35.1s
35.1s
Quantization
BF16
—
Harness
openrouter
custom

The race

1 lead change

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

Apodex 1.1 Mini (free)Clover 35B A3B Preview
0255075100501001502530

Every test case

153 test cases

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

11 won by Apodex 1.1 Mini (free)21 won by Clover 35B A3B Preview121 tied