MiniMax M3
by MiniMax · minimax/minimax-m3 · #125 cheapest of 325 paid models
Prices updated Jul 22, 2026, 1:31 AM UTC · refreshed hourly
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,...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.30
per 1M tokens
Output
$1.20
per 1M tokens
Blended (3:1)
$0.525
3 input : 1 output
Cache read
$0.06
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for MiniMax M3 with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.525 | $0.0318 | — |
| 50% | $0.435 | $0.0198 | −38% |
| 90% | $0.363 | $0.0102 | −68% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.00051 | $0.51 |
| Document summary | 8,000 | 1,000 | $0.0036 | $3.60 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.0114 | $11.40 |
| Long-context analysis | 150,000 | 5,000 | $0.051 | $51.00 |
Performance
Intelligence Index
44.4
AA composite quality
Coding Index
58.6
Math Index
—
Agentic Index
35.4
Output speed
97 tok/s
median
Time to first token
1.29s
median
Time to first answer
21.90s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| GPQA Diamond | 92.9% |
| Humanity's Last Exam | 37.1% |
| SciCode | 45.4% |
| IFBench | 82.9% |
| AA-LCR (Long Context Reasoning) | 74.0% |
| Terminal-Bench Hard | 42.4% |
| Terminal-Bench 2.1 | 65.2% |
| τ²-Bench (Telecom) | 88.9% |
| τ³-Bench Banking | 13.0% |
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,287 |
| Web apps | 1,263 |
| Full-stack | 1,250 |
| Mobile apps | 1,256 |
| Android native | 1,099 |
| UI components | 1,280 |
| Data viz | 1,277 |
| SVG | 1,227 |
| 3D | 1,280 |
| Game dev | 1,280 |
| Code categories | 1,287 |
| ASCII art | 1,198 |
Specs
Context window
1.05M
tokens
Max output
512K
tokens
Input modalities
text, image, video
Tokenizer
Other