[mle]/DaBench

Apodex 1.1 Mini (free) vs MiniMax M3.1 Flash

Side A
Model page
Side B
Model page
Overall
25
vs
Overall
35

MiniMax M3.1 Flash is clearly better. It wins 41 of the 57 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. Apodex 1.1 Mini (free) spends 3.7× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
2535.1s
8.5K, 75% thinking2.3K, 73% thinking
3521.3s
Agents
2639.4s
6.2K, 69% thinking3.1K, 71% thinking
3637.0s
Structured Extraction
5227.2s
7.7K, 68% thinking1.1K, 74% thinking
487.96s
Coding
446.5s
12.1K, 94% thinking4.6K, 66% thinking
5840.6s
Grounded Answers
2926.0s
6.9K, 70% thinking1.2K, 71% thinking
339.75s
Instruction Following
3644.5s
11.4K, 79% thinking1.7K, 70% thinking
5314.7s
Business Logic
1435.4s
10K, 56% thinking3.5K, 94% thinking
2425.9s
Security & permissions
4027.3s
5.8K, 76% thinking1.3K, 60% thinking
3012.7s
Diagnosis & investigation
034.6s
5.1K, 84% thinking2.1K, 51% thinking
022.8s

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)
135 tok/s
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
256K
1M (better)
Cost per run
Not measured
Not measured
Cost per case
Not measured
Not measured
Time per case
35.1s
21.3s (better)
Quantization
BF16
—
Harness
openrouter
openrouter

The race

No 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)MiniMax M3.1 Flash
025507510050100150Perfect run2535

Every test case

153 test cases

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

16 won by Apodex 1.1 Mini (free)41 won by MiniMax M3.1 Flash96 tied