MiniMax-M2.7 vs MiniMax-M3
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by MiniMax
|
by MiniMax
|
|---|---|---|
| Overview | ||
| Status | Current budget Open weights | Current budget Open weights |
| Released | Oct 1, 2025 | Jun 1, 2026 |
| Pricing per million tokens | ||
| Input | $0.300/Mtok | $0.300/Mtok |
| Output | $1.20/Mtok | $1.20/Mtok |
| Blended avg | $0.750/Mtok | $0.750/Mtok |
| Cached input | $0.060/Mtok | $0.060/Mtok |
| Specifications | ||
| Context window | 205K tokens | 1M tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
text
|
textimagevideo
|
| Benchmarks sources: LMSYS Chatbot Arena (UC Berkeley) / Vellum LLM Leaderboard | ||
| Independent composite scores measured for both models — each on its own scale | ||
| AA Intelligence Index | 38.1 | 44.4 |
| AA Agentic Index | 25.6 | 35.4 |
| AA-Omniscience Index (−100–100) | 0.7 | 1.4 |
| LMSYS Chatbot Arena | 1416.9 ELO | 1445.3 ELO |
| GDPval-AA v2 | — | 1390.1 ELO |
| Per-benchmark results published for MiniMax-M3 only — no independent per-benchmark scores exist for MiniMax-M2.7 yet | ||
| Avg benchmark score | — | 63.1 |
| Perf / dollar | — | 84.1 |
| GPQA Diamond | — | 93 |
| SWE-Bench Verified | — | 45 |
| Terminal-Bench 2.1 | — | 66 |
| MCP Atlas | — | 74.2 |
| BrowseComp | — | 83.5 |
| OSWorld-Verified | — | 70.1 |
| Humanity's Last Exam | — | 22 |
| ARC-AGI 2 | — | 20 |
| AIME 2025 | — | 55 |
| MMMLU | — | 76 |
| BFCL | — | 65 |
| HumanEval | — | 80.5 |
| MATH 500 | — | 70 |
| Providers | ||
| Available from |
MiniMax — $0.300/$1.20/Mtok
|
MiniMax — $0.300/$1.20/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | MiniMax-M2.7 | MiniMax-M3 | Savings |
|---|---|---|---|
| 1M tokens | $0.75 | $0.75 | $0 (0%) |
| 10M tokens | $7.5 | $7.5 | $0 (0%) |
| 100M tokens | $75 | $75 | $0 (0%) |
| 1000M tokens | $750 | $750 | $0 (0%) |
When to pick which
distilled from the pricing and spec data above- this pairing gives MiniMax-M2.7 no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
- you need the longer context: 1M tokens vs 205K (4.9×)
Price movement
recorded changes since we began tracking each model · last checked Aug 1, 2026| Model | Changes | Latest move | Old → new $/M |
|---|---|---|---|
| MiniMax-M2.7 | 0 | — no move | Stable since we began tracking — no change recorded |
| MiniMax-M3 | 1 | price cut on Jun 17, 2026 | $0.6/$2.4 → $0.3/$1.2/Mtok |
The cheaper option here — MiniMax-M2.7 costs $0.750/Mtok blended on MiniMax.
Summary
MiniMax-M2.7 by MiniMax costs $0.300/Mtok input and $1.20/Mtok output, with a 205K-token context window. It supports text input.
MiniMax-M3 by MiniMax costs $0.300/Mtok input and $1.20/Mtok output, with a 1M-token context window. It supports text, image, video input.
The two aren't directly comparable on average benchmark score: MiniMax-M3 has published per-benchmark results, while MiniMax-M2.7 does not yet — it is measured today on independent composites (see the table above).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.