minimax Open weights Vision Reasoning

MiniMax M3 pricing & benchmarks

Released May 31, 2026 · served by 8 providers · ranked #20 by intelligence in our index

Input / 1M tokens$0.240
Output / 1M tokens$0.960
Context window1.0M
Cached input$0.060
Intelligence index44.4
Coding index58.6
Agentic index35.4
1M in + 300K out$0.53

What MiniMax M3 is

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...

Providers and prices for MiniMax M3

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
GMICloudcheapest $0.240 $0.960 1.0M fp8 98.5%
Novita $0.300 $1.20 1M fp8 98.7%
Venice $0.300 $1.20 524K fp8 95.3%
Minimax $0.300 $1.20 524K fp8 99.1%
AtlasCloud $0.300 $1.20 524K fp8 98.3%
Together $0.300 $1.20 524K 97.1%
Morph $0.300 $1.20 256K 93.5%
DeepInfra $0.300 $1.20 524K fp8 97.7%

MiniMax M3 benchmark results

Design Arena head-to-head results, as reported through the OpenRouter model API.

CategoryArenaRankEloWin rate
mobileapps agents #4 1263 57.5%
agenticgamedev agents #7 1201 52.9%
fullstack agents #9 1250 53%
htmlslides agents #12 1202 50.7%
webapps agents #14 1246 51.6%
svg models #19 1223 50.4%
androidnative agents #20 1120 35.3%
website models #20 1286 54.8%
codecategories models #21 1285 54.5%
asciiart models #22 1193 47.3%
dataviz models #23 1265 53.6%
uicomponent models #23 1278 53.2%
3d models #24 1279 54.6%
gamedev models #26 1274 50.5%

Cheaper models in the same class

Models scoring within 4 points of MiniMax M3 on the Intelligence Index, but with a lower output price.

Compare MiniMax M3 head-to-head

FAQ

How much does the MiniMax M3 API cost?

$0.240 per 1M input tokens and $0.960 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.53.

Is MiniMax M3 free?

No. It is a paid model starting at $0.240 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for MiniMax M3?

GMICloud at $0.240 input / $0.960 output per 1M tokens (fp8 quantization).

Can I self-host MiniMax M3?

Yes — weights are published as MiniMaxAI/Minimax-M3 on Hugging Face, so you can run it on your own hardware.