← Model list
MiniMax-M3
minimax·minimax/MiniMax-M3·GA·Open weights·minimax series·NEW
MiniMax multimodal model for long-context coding, perception, and agent planning
At a glance
Official price$0.30 / $1.20 per 1M
Cache read $0.06 · MiniMax (minimaxi.com)
Lowest paid$0.225 / $0.90
EmpirioLabs AI Gateway · 2.7× spread
5 more $0 channels
Context1,000,000
⚠ Providers report 262,144–1,048,576; the table below is authoritative
Output limit128,000
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImageVideoPDF
Knowledge cutoff2025-01
Released / updated2026-06-01 / 2026-06-01
Quality & performanceArtificial Analysis · Intelligence Index v4.1
Intelligence45.4
Value115
Coding58.6
Agentic36.1
Output speed90 tok/s
TTFT1.35 s
Cost per task$0.1387
Value formulaIQ 45.4 ÷ min blended $0.394 = 115
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
Available at 38 providers32 with public prices · 2 free · 3 subscription-covered
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| Nvidia minimaxai/minimax-m3 | First-party | Free | — | — | 1,000,000 | 16,384 | ||
| Kenari minimax-m3 | Gateway | Free | — | — | 512,000 ⚠ | 128,000 | ||
| MiniMax Token Plan (minimaxi.com) | Gateway | Subscription | $0 | $0 | 1,000,000 | 128,000 | ||
| MiniMax Token Plan (minimax.io) | Gateway | Subscription | $0 | $0 | 1,000,000 | 128,000 | ||
| SCNet Token Plan | Gateway | Subscription | $0 | — | 512,000 ⚠ | 128,000 | ||
| EmpirioLabs AI minimax-m3 | Gateway | $0.225 tiered >512K: $0.45 | $0.90 | $0.045 | — | 1,000,000 | 524,288 | |
| Weights & Biases | Cloud | $0.23 | $0.96 | $0.05 | — | 262,144 ⚠ | 262,144 | |
| Requesty minimax-m3 | Gateway | $0.24 | $0.96 | $0.048 | — | 1,000,000 | 128,000 | |
| Deep Infra | Cloud | $0.28 | $1.10 | $0.056 | — | 524,288 ⚠ | 128,000 | |
| Lilac minimaxai/minimax-m3 | Gateway | $0.28 | $1.10 | $0.05 | — | 1,048,576 ⚠ | 1,048,576 | |
| MiniMax (minimaxi.com)Official | First-party | $0.30 tiered >512K: $0.60 | $1.20 | $0.06 | — | 1,000,000 | 128,000 | |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Your usage cost
1EmpirioLabs AI$68.40
2Weights & Biases$72.40
3Requesty$72.96
4Lilac$83.40
5Deep Infra$84.12
6NanoGPT$91.20
12MiniMax (minimaxi.com) · Official$91.20
5 more channels offer $0 (Nvidia, Kenari, MiniMax Token Plan (minimaxi.com) etc.); free tiers usually have rate limits and no SLA, excluded from ranking.
Switch to EmpirioLabs AI to save $22.80/mo (25%)
Note: this is a gateway; verify availability and rate limits yourself.
Benchmark6 items
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Verified | harness: Claude Code | 80.5 | resolved | Source ↗ |
| SWE-Bench Pro | harness: Claude Code | 59 | resolve rate | Source ↗ |
| Terminal-Bench | v2.1 | 66 | success rate | Source ↗ |
| BrowseComp | — | 83.52 | accuracy | Source ↗ |
| MCP Atlas | — | 74.2 | success rate | Source ↗ |
| OSWorld-Verified | — | 70.06 | success rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Reasoning control
Toggle (on / off)effort = noneeffort = loweffort = mediumeffort = higheffort = maxbudget_tokens ≥ 1 ≤ 128,000effort = xhigheffort = minimal
13 / 38 providers expose no reasoning control (reasoning_options: []).
Related models
MiniMax H3same series—MiniMaxAI/MiniMax-M2.5same series$0.30 / $1.20MiniMax M3 Thinkingsame series$0.30 / $1.20DeepSeek V4 Flashcheaper alternative$0.14 / $0.28MiMo-V2.5cheaper alternative$0.14 / $0.28Gemini 2.5 Flash-Litecheaper alternative$0.10 / $0.40Qwen Pluscheaper alternative$0.115 / $0.287
Price historyone sample accumulated per data sync
Input listOutput listMin blended