AI Battle > Rankings > MiniMax-M3

MiniMax-M3 — Placements in 57 Rankings (2026)

MiniMax-M3, published by MiniMax, appears in 57 of the 129 datasets AI Battle tracks. Its strongest published placement is #1 of 20 in AA-Omniscience Hallucination Rate.

Last updated: . Every placement below comes from the named public dataset; AI Battle joins them by slug and changes nothing else.

All placements for MiniMax-M3

RankingRankValue
AA-Omniscience Hallucination Rate#1 of 200.18
𝜏³-Banking: Output Tokens per Task#1 of 202468
Pricing: Cache Hit, Input, and Output#4 of 200.30
Pricing: Cache Hit, Input, and Output#4 of 200.30
Humanity's Last Exam: Cost per Task#4 of 200.00
AA-LCR v1.1: Cost per Task#4 of 200.03
Context Window#5 of 201000000
𝜏³-Banking: Cost per Task#5 of 200.00
AA-Omniscience Index: Cost Breakdown#6 of 200.22
AA-LCR v1.1: Output Tokens per Task#6 of 20160
𝜏³-Banking: Time per Task#6 of 201.06
MMMU-Pro: Total Cost to Run#6 of 190.41
End-to-End Response Time#7 of 201.29
Cost to Run Artificial Analysis Intelligence Index#7 of 2035.0
Latency: Time To First Answer Token#7 of 201.29
Cost to Run Artificial Analysis Intelligence Index#7 of 2035.0
AA-LCR v1.1: Time per Task#7 of 200.22
GDPval-AA v2: Cost per Task#7 of 200.00
Cost per Intelligence Index Task#8 of 200.01
Cost per Intelligence Index Task#8 of 200.01
AA-Briefcase Cost per Task#8 of 200.10
Mean Turns per Task#9 of 20101
SciCode: Cost per Task#9 of 200.01
Output Speed#10 of 20107
Output Speed#10 of 20107
Terminal-Bench v4.0: Cost per Task#10 of 200.00
Time per Task#10 of 2016.0
AA-Briefcase Output Tokens per Task#11 of 2081849
GDPval-AA v2: Average Turns per Task#11 of 2054.2
Model Size: Total and Active Parameters#8 of 1423.0
Output Tokens per Intelligence Index Task#12 of 2026374
Time per Intelligence Index Task#12 of 207.67
Time per Intelligence Index Task#12 of 207.67
Output Tokens per Intelligence Index Task#12 of 2026374
GDPval-AA v2: Output Tokens per Task#12 of 2031654
Humanity's Last Exam: Output Tokens per Task#13 of 20466
AA-Briefcase Tool Calls Breakdown, Avg per Task#13 of 2051.2
AA-LCR v1.1: Score#14 of 200.83
MMMU-Pro: Score#14 of 190.79
AA-Omniscience Index#15 of 201.35
AA-Omniscience Index#15 of 201.35
Artificial Analysis Openness Index: Components#9 of 124.00
Humanity's Last Exam: Time per Task#15 of 203.42
Artificial Analysis Openness Index: Score#10 of 1233.3
MMMU-Pro: Token Usage#16 of 191359529
Artificial Analysis Finance & Accounting Index#18 of 2030.7
AA-Omniscience Index: Token Usage#18 of 20716865
AA-AnalystAgent pass^5#10 of 110.10
AA-Briefcase Elo#19 of 201096
Artificial Analysis Intelligence Index by Open Weights / Proprietary#19 of 2029.6
GDPval-AA v2 Leaderboard#19 of 201304
Artificial Analysis Intelligence Index by Open Weights / Proprietary#19 of 2029.6
Artificial Analysis Intelligence Index#19 of 2029.6
AA-Briefcase Elo#19 of 201096
Artificial Analysis Intelligence Index#19 of 2029.6
Terminal-Bench v4.0: Output Tokens per Task#20 of 2098158
AA-Omniscience Index: Score#20 of 201.35

Price vs intelligence

MiniMax-M3intelligence index →input price / 1M → $4.00
Each dot is a model, joined by AI Battle from two public datasets — the intelligence index ranking and the published input pricing ranking. Every published value is shown as-is, no averaging. The dashed line is the value frontier: no model above or to the right of it offers a lower input price at its intelligence level.

How to read this profile

Each row is a separate measurement in a separate dataset — the position reflects the metric of that ranking, and a higher number is not "better" for every dataset (times and prices are costs). AI Battle deliberately does not average placements into a single score: datasets differ in coverage, so a total would hide more than it shows.

Raw data

Machine-readable profile: /data/models/minimax-m3.json.

Method

Placements, joins and this page are computed by AI Battle from public datasets — see the methodology for sources, limitations and corrections.