[mle]/DaBench

DeepSeek V4.1 Flash 1017 vs MiniMax M3.1 Flash

Side A
Model page
Side B
Model page
Overall
56
vs
Overall
35

DeepSeek V4.1 Flash 1017 is clearly better. It wins 53 of the 70 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. DeepSeek V4.1 Flash 1017 spends 2.8× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
5626.0s
5.3K, 91% thinking2.3K, 73% thinking
3521.3s
Agents
5826.7s
3.3K, 71% thinking3.1K, 71% thinking
3637.0s
Structured Extraction
8117.2s
4.2K, 93% thinking1.1K, 74% thinking
487.96s
Coding
4840.8s
9.4K, 95% thinking4.6K, 66% thinking
5840.6s
Grounded Answers
7119.8s
4.4K, 91% thinking1.2K, 71% thinking
339.75s
Instruction Following
7625.6s
5.5K, 91% thinking1.7K, 70% thinking
5314.7s
Business Logic
6726.6s
6.6K, 97% thinking3.5K, 94% thinking
2425.9s
Security & permissions
4826.6s
4.7K, 87% thinking1.3K, 60% thinking
3012.7s
Diagnosis & investigation
020.2s
3.2K, 74% 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
243 tok/s (better)
135 tok/s
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
1M
1M
Cost per run
$0.75
Not measured
Cost per case
$0.0049
Not measured
Time per case
26.0s
21.3s (better)
Quantization
—
—
Harness
openrouter
openrouter

The race

3 lead changes

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

DeepSeek V4.1 Flash 1017MiniMax M3.1 Flash
025507510050100150Perfect run5635

Every test case

153 test cases

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

53 won by DeepSeek V4.1 Flash 101717 won by MiniMax M3.1 Flash83 tied