MiniMax M1 40k
Jun 2025
MiniMax M1 40kCompare inon
Benchmarks of MiniMax M1 40k8
BenchmarkByScoreRank
Overall1
Knowledge1
Science1
Coding1
Legacy · saturated or no longer reported4
No benchmark matches
—per task?
self-host onlyprice
—output speed?
1Mcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGemma 4 E4BthinkingOpen+7%—$0.008Context: 1M → 128KOpenMiniMax M3thinking×2—$0.51Context: 1M → 524K+2
BetterGPT OSS 20BhighOpen+58%—$0.011Context: 1M → 131K
BetterMistral Small 4thinkingOpen+17%—$0.015Context: 1M → 262K
BetterNemotron 3 Nano 30B A3BthinkingOpen+72%—$0.017Context: 1M → 131KModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
Gemma 4 E4BthinkingOpen+7%—$0.008Context: 1M → 128KGPT OSS 20BhighOpen+58%—$0.011Context: 1M → 131KMistral Small 4thinkingOpen+17%—$0.015Context: 1M → 262KNemotron 3 Nano 30B A3BthinkingOpen+72%—$0.017Context: 1M → 131K
MiMo v2.5thinkingOpen+63%—$0.019ModelScoreCostPer taskWhat changes
MiniMax M3thinkingOpen×2—$0.51Context: 1M → 524K+2Nemotron 3 UltrathinkingOpen+97%—$0.58Memory (4-bit): 274 GB → 336 GBNemotron Cascade 2 30B A3BOpen+95%——MiMo v2.5 ProthinkingOpen+94%—$0.054Memory (4-bit): 274 GB → 614 GB
Qwen3.5 397B A17BthinkingOpen+91%—$0.47Context: 1M → 262KModelScoreCostPer taskWhat changes
MiniMax M3thinkingOpen×2—$0.51Context: 1M → 524K+2MiniMax M2.7thinkingOpen+84%—$0.30 / $1.20Context: 1M → 205KMiniMax M2thinkingOpen+75%—$0.30 / $1.20Context: 1M → 205KMiniMax M2.5thinkingOpen+74%—$0.30 / $1.20Context: 1M → 205KMiniMax M2.1thinkingOpen+69%—$0.30 / $1.20Context: 1M → 205KMiniMax M1 80kOpensame—$0.55 / $2.20MiniMax M3.1 Flashthinkingn/a—$0.22Complete model response must be appendedMiniMax M1unstatedOpenn/a—$0.40 / $2.20No strict JSONAll MiniMax models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✗ 80 GB GPU
LicenceApache-2.0free for commercial use
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam7.8#211
AA Intelligence Index 4.3.2≈10#207
GPQA Diamond68.2#195
AIME 2025 · AA archived13.7#123
Coding
LiveCodeBench · AA archived65.7#60
Agents & Tool Use
τ²-bench · AA31.6#146
Writing & Language
IFBench (instructions)41.2#154
FAQ
How much does MiniMax M1 40k cost?›
There is no per-unit API price: self-host only.
Is MiniMax M1 40k free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is MiniMax M1 40k, and what is it best at?›
#210 of 334 in Chat & Knowledge, #181 of 239 in Coding, #144 of 242 in Agents & Tool Use, #153 of 245 in Writing & Language.
Is MiniMax M1 40k better than ERNIE 5.0 Thinking Preview?›
On IFBench (instructions), MiniMax M1 40k scores 41.2 vs 41.4 — about the same.
What is the context window of MiniMax M1 40k?›
1M tokens.
When was MiniMax M1 40k released? Is there a newer version?›
Released 2025-06-17. It has been superseded by MiniMax M2.
Is MiniMax M1 40k open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run MiniMax M1 40k locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can MiniMax M1 40k read images?›
No, it accepts text only.
What is MiniMax M1 40k?›
MiniMax M1 40k is an open-weights language model from MiniMax, released Jun 2025.
What are the best alternatives to MiniMax M1 40k?›
Gemma 4 E4B (better), MiniMax M3 (open weights).