GPT-6 Luna vs MiMo-V2.6-Pro, head to head

Side A

Side B

Overall

GPT-6 Luna is clearly better. It wins 42 of the 59 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. MiMo-V2.6-Pro spends 3.8× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAGPT-6 LunaBMiMo-V2.6-ProGPT-6 LunaMiMo-V2.6-Pro

Overall

6149.1s

1.9K5.9K

482m 13s

Agents

Runs multi-step jobs through tools, within a written policy

561m 18s

1.3K8.7K

474m 48s

Structured Extraction

Turns logs, diffs and records into exact JSON

9514.5s

9993.4K

601m 26s

Coding

Writes code and SQL that pass hidden tests

732m 18s

3.6K10.7K

242m 58s

Grounded Answers

Answers from the given documents with sources, never guesses

6715.6s

1.1K3.2K

621m 9s

Instruction Following

Applies every rule, including ones changed mid-conversation

8432.3s

2K4.5K

591m 23s

Business Logic

Computes billing, scheduling and pricing rules exactly

3743.1s

3.2K7.1K

571m 52s

Security & permissions

Ignores planted instructions, still does the allowed work

6224.5s

1K3.6K

711m 41s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1737.7s

1.1K6.2K

02m 54s

What you'd run

The build behind each score. Orange marks the faster, smaller, cheaper or longer-context side.

SpecAGPT-6 LunaBMiMo-V2.6-Pro

Speed

90 tok/s

55 tok/s

Machine

Hosted

Memory at 32K

—

Context

1.1M

Cost per run

$0.10

$1.80

Cost per case

$0.00067

$0.012

Time per case

49.1s

2m 13s

Quantization

—

BF16

Harness

openrouter

The race

3 lead changes

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

GPT-6 LunaMiMo-V2.6-Pro

Every test case

153 test cases

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

Agents54

Structured Extraction91

Coding130

Grounded Answers32

Instruction Following71

Business Logic37

Security & permissions12

Diagnosis & investigation10

42 won by GPT-6 Luna17 won by MiMo-V2.6-Pro94 tied