Pass IndexThe State of AISign in

MiniMax M3

MiniMax-M3 is a frontier-class foundation model that unites the three capabilities defining today's frontier: a 1M-token context window, frontier coding and agentic performance, and native multimodality — the first open-weight model to deliver all three in a single system.

text + image + file → text · made by MiniMax

$0.3$1.2per Mtok in / out
MiniMax · 14 sellers

Sold by 32 ways

SellerLaneRate
MiniMax apistandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedplatform.minimax.io · read 2026-08-25
MiniMax apipriority$0.45per Mtok in$1.8per Mtok out$0.09per Mtok cachedplatform.minimax.io · read 2026-08-25
MiniMax apilong-context$0.6per Mtok in$2.4per Mtok out$0.12per Mtok cachedplatform.minimax.io · read 2026-08-25
MiniMax apipriority long-context$0.9per Mtok in$3.6per Mtok out$0.18per Mtok cachedplatform.minimax.io · read 2026-08-25
Nous Research 2 ways$0.23per Mtok in$0.96per Mtok out$0.05per Mtok cached
Nous Research aggregatorstandard$0.23per Mtok in$0.96per Mtok out$0.05per Mtok cachedinference-api.nousresearch.com · read 2026-09-12
Nous Research aggregatorbatch$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedinference-api.nousresearch.com · read 2026-09-12
OpenRouter 7 ways$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cached
OpenRouter aggregatorcoreweave/fp4$0.23per Mtok in$0.96per Mtok out$0.05per Mtok cachedopenrouter.ai · read 2026-08-25
OpenRouter aggregatorgmicloud/fp8$0.24per Mtok in$0.96per Mtok out$0.048per Mtok cachedopenrouter.ai · read 2026-09-10
OpenRouter aggregatornot on the seller's list todaymorph/fp4$0.26per Mtok in$1.02per Mtok outopenrouter.ai · read 2026-08-25
OpenRouter aggregatordeepinfra/fp8$0.28per Mtok in$1.1per Mtok out$0.056per Mtok cachedopenrouter.ai · read 2026-08-25
OpenRouter aggregatorstandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedopenrouter.ai · read 2026-08-24
OpenRouter aggregatormara$0.6per Mtok in$2.4per Mtok outopenrouter.ai · read 2026-09-15
OpenRouter aggregatorsambanova$0.6per Mtok in$2.4per Mtok outopenrouter.ai · read 2026-08-25
OpenRouter aggregatormodelrun/fp4$0.75per Mtok in$3per Mtok out$0.15per Mtok cachedopenrouter.ai · read 2026-08-25
DeepInfra aggregatorstandard$0.28per Mtok in$1.1per Mtok out$0.056per Mtok cachedapi.deepinfra.com · read 2026-08-25
Hugging Face Inference Providers 5 ways$0.3per Mtok in$1.2per Mtok out
Hugging Face Inference Providers aggregatordeepinfra$0.28per Mtok in$1.1per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatorstandard$0.3per Mtok in$1.2per Mtok outmodels.dev · read 2026-09-16
Hugging Face Inference Providers aggregatorfireworks-ai$0.3per Mtok in$1.2per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatornovita$0.3per Mtok in$1.2per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatortogether$0.3per Mtok in$1.2per Mtok outrouter.huggingface.co · read 2026-08-25
ElectronHub aggregatorstandard$0.3per Mtok in$1.2per Mtok outapi.electronhub.ai · read 2026-09-16
Fireworks AI 2 ways$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cached
Fireworks AI aggregatorstandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cacheddocs.fireworks.ai · read 2026-08-24
Fireworks AI aggregatorpriority$0.45per Mtok in$1.8per Mtok out$0.09per Mtok cachedraw.githubusercontent.com · read 2026-09-16
Nebius Token Factory aggregatorstandard$0.3per Mtok in$1.2per Mtok outtokenfactory.nebius.com · read 2026-08-25
Novita AI aggregatorstandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedapi.novita.ai · read 2026-08-25
Requesty 3 ways$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cached
Requesty aggregatorfireworks$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedrouter.requesty.ai · read 2026-08-28
Requesty aggregatorminimaxi$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedrouter.requesty.ai · read 2026-08-28
Requesty aggregatortensorx$0.4per Mtok in$2per Mtok out$0.1per Mtok cachedrouter.requesty.ai · read 2026-08-28
Together AI aggregatorstandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cacheddocs.together.ai · read 2026-08-24
Vercel AI Gateway aggregatorstandard$0.3per Mtok in$1.2per Mtok out$0.06per Mtok cachedai-gateway.vercel.sh · read 2026-08-25
SambaNova Cloud aggregatorstandard$0.6per Mtok in$2.4per Mtok outapi.sambanova.ai · read 2026-08-31
PPIO aggregatorstandard$4.2per Mtok in$16.8per Mtok outapi.ppinfra.com · read 2026-09-16

Measured 39 standings

PlaceBoardMetricScore
11thof 110OSWorld-Verified Results (All)Success Rate (Avg±Std)75.2
11thof 17SWE-rebench Leaderboard (default window 15/05/2026–01/07/2026, 111 problems)Resolved Rate (%)47.2
14thof 16Osworld 2 — Epoch AIBinary accuracy0.046
15thof 26Surface evolver bench — Epoch AIMean score0.55
16thof 36Design Arena · htmlslides (agents)Elo1182
16thof 23Gbaeval — Epoch AIOverall score0.009
21stof 43Design Arena · python-pptxslides (agents)Elo1207
26thof 30Frontiercode — Epoch AIMain score0.147
27thof 81Forecastbench — Epoch AIOverall score60.4
31stof 64Design Arena · fullstack (agents)Elo1207
31stof 62Design Arena · webapps (agents)Elo1215
31stof 313GPQA Diamondmean_score90.909
32ndof 40Design Arena · agenticgamedev (agents)Elo1159
32ndof 63Design Arena · mobileapps (agents)Elo1193
37thof 55Design Arena · androidnative (agents)Elo1167
38thof 64Proofbench — Epoch AIAccuracy18
42ndof 60Vending bench 2 — Epoch AIScore2157.772
43rdof 162Design Arena · website (models)Elo1271
45thof 121Webdev arena — Epoch AIArena Score1487.64
47thof 156Design Arena · codecategories (models)Elo1265
49thof 143Artificial Analysis · Agentic IndexIndex30.8
49thof 92Design Arena · asciiart (models)Elo1180
49thof 150Design Arena · uicomponent (models)Elo1260
50thof 109Design Arena · svg (models)Elo1194
51stof 147Design Arena · 3d (models)Elo1239
52ndof 152Design Arena · dataviz (models)Elo1249
54thof 155Design Arena · gamedev (models)Elo1237
58thof 123Lmca — Epoch AIScore33.7
64thof 102Simplebench — Epoch AIScore (AVG@5)0.458
65thof 184Artificial Analysis · Coding IndexIndex58.6
70thof 161Dtbench — Epoch AIAccuracy78.93
72ndof 302Artificial Analysis · Intelligence IndexArtificial Analysis Intelligence Index30
79thof 112Ale bench — Epoch AIPerformance640.02
87thof 177Critpt — Epoch AIAccuracy3.714
89thof 168Scicode — Epoch AIScore45.37
108thof 127Mystery game puzzles — Epoch AImean_score (none)8
109thof 222Chess puzzles — Epoch AImean_score14
126thof 290OTIS Mock AIME 2024-2025mean_score71.111
284thof 656Epoch capabilities index — Epoch AIECI Score146.57

About

MiniMax M3 — a text model from MiniMax, sold by 14 companies from $0.23 in and $0.96 out per million tokens, placed 11th of 110 on OSWorld-Verified Results (All).

It takes text, images and files and returns text, with a context window of 512,000 tokens. It was published in May 2026, trained on material up to January 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Fourteen companies sell it. The cheapest is $0.23 in and $0.96 out per million tokens at Nous Research. Beside the standard rate there are faster, separately routed, batch and regional lanes. It has been measured on 39 boards, and stands best at 11th of 110 on OSWorld-Verified Results (All).

Every current figure

Maker
MiniMax
Register
model
Takes
text + image + file
Returns
text
Context
512,000 tokens
Published
May 2026
Knowledge to
January 2025
Parameters
427 billion
Licence
not read
Sellers
14
Price
$0.23 in and $0.96 out per million tokens — Nous Research
Boards
39
Best place
11th of 110 — OSWorld-Verified Results (All)

Known as 8 names

MiniMaxAI/MiniMax-M3fireworks/minimax-m3MiniMax-M3minimax/minimax-m3minimax/minimax-m3:batchminimaxi/minimax-m3tensorx/minimax-m3MiniMax: MiniMax M3 (batch)