[mle]/DaBench

GLM 5.3 Flash vs MiniMax M3.1 Flash

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

Too close to call. MiniMax M3.1 Flash wins 33 of the 61 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. GLM 5.3 Flash spends 1.5× as many thinking tokens per test case.

Thinking tokensAnswer tokens
TaskAB
Overall
3555.5s
2.8K, 89% thinking2.3K, 73% thinking
3521.3s
Agents
1037.0s
607, 81% thinking3.1K, 71% thinking
3637.0s
Structured Extraction
4847.2s
2.7K, 93% thinking1.1K, 74% thinking
487.96s
Coding
541m 52s
5.3K, 83% thinking4.6K, 66% thinking
5840.6s
Grounded Answers
3756.8s
2.9K, 93% thinking1.2K, 71% thinking
339.75s
Instruction Following
4340.8s
2K, 77% thinking1.7K, 70% thinking
5314.7s
Business Logic
331m 18s
5.8K, 98% thinking3.5K, 94% thinking
2425.9s
Security & permissions
5525.2s
963, 81% thinking1.3K, 60% thinking
3012.7s
Diagnosis & investigation
027.0s
563, 63% 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
60 tok/s
135 tok/s (better)
Machine
Hosted
Hosted
Memory at 32K
—
—
Context
1M
1M
Cost per run
$0.19
Not measured
Cost per case
$0.0013
Not measured
Time per case
55.5s
21.3s (better)
Quantization
FP8
—
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.

GLM 5.3 FlashMiniMax M3.1 Flash
025507510050100150Perfect run3535

Every test case

153 test cases

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

28 won by GLM 5.3 Flash33 won by MiniMax M3.1 Flash92 tied