Model result, rank 18 of 27

MiniMax M3

OpenRouter · extra-high reasoning. Public result card with the model’s overall score, lane measurements, runtime and cost telemetry, and the ranking formula.

Compare all models Download public JSON

Figure 1 Lane profile against the cohort Each bar is one measured lane on the shared 0 to 100 scale. The thin mark is the cohort median; the bright mark is the best score any ranked model reached on that lane.

Full / Agenticlane rank #15 of 27

77.92

SWE MVPlane rank #6 of 27

86.88

Hard Intelligencelane rank #21 of 27

66.77

MiniMax M3
Cohort median
Cohort best

Overall score77.19

Rank #18

Full / Agentic77.92

Full rank #15

SWE MVP86.88

SWE rank #6

Hard Intelligence66.77

Hard rank #21

Measured cost$0.182

100.0% reliability

Overall

All-around publication view

Score77.19

Formulamean(Full, SWE, Hard Intelligence)

BasisOpenRouter · extra-high reasoning

The overall score averages the measured major lanes while keeping each source measurement visible.

Lane 01

Full / Agentic benchmark

Final77.92

Capability92.40

Agentic83.83

Pass rate85.7%

Prompts43

This lane captures instruction following, structured behavior, tool discipline, and general agentic reliability.

Lane 02

Software engineering MVP

SWE score86.88

Focused final75.84

Capability86.15

Daily driver73.36

Prompts24

This lane is closer to implementation usefulness: source handling, architecture cleanliness, and deliverable quality.

Lane 03

Hard Intelligence diagnostic

Hard score66.77

Active inquiry17.50

Online adaptation97.50

Self-repair82.08

Authority integrity70.00

Hard Intelligence measures active inquiry, online adaptation, evidence-driven self-repair, and authority/salience integrity.

Telemetry

Runtime economics

Total cost$0.182

Cost / scored item$0.0024

Seconds / timed item27.52s

Runtime coverage100.0%

Recorded tokens / item6.7k

Token coverage100.0%

Cost, time, and token basis are normalized telemetry. They explain tradeoffs; they do not overwrite the capability score yet.

Interpretation

Why the result lands here.

The model is stronger in the SWE lane than in the Full/Agentic lane; the overall score is therefore shown with both component lanes visible. Hard Intelligence score is 66.77 and contributes to the overall score alongside Full/Agentic and SWE. Hard Intelligence is published alongside Full/Agentic and SWE for the current ranking.