Your sovereign, private MiniMax 2.7
MiniMax M2.7 is an interleaved-thinking MoE reasoner (~230B total / 10B active) tuned for agentic workflows: multi-step tool calling, long-horizon task completion and cost-efficient inference with a large usable context. A reasoning model in the large tier, served on ToothFairyAI's AU, EU and US endpoints — private by default, billed in US dollars per million tokens at the published rates below. No training on your data, no vendor lock-in.
How MiniMax 2.7 compares
Independent intelligence index for MiniMax 2.7 against every peer in the Large tier (and the tier above where needed) — same source, same tests.
Latency — end-to-end seconds for a 500-token answer
Capability indexes
Domain-weighted agentic performance for MiniMax 2.7.
Capability index vs large peers
The evaluations behind the index
Independent benchmark results for MiniMax 2.7 — Elo scores as published; pass rates shown as percentages.
Per-evaluation economics
| Evaluation | Score | Time / task | Output tokens / task |
|---|---|---|---|
| AA-Briefcase | 712.77 | 8.1 min | 31863 |
| GDPval-AA | 998.64 | 6.1 min | 24059 |
| AutomationBench-AA | 4.4% | 2.6 min | 10312 |
| Terminal-Bench 4.0 | 0.0% | 12.5 min | 49112 |
| SciCode | 50.1% | 1.5 min | 6036 |
| Humanity's Last Exam | 29.6% | 6.4 min | 25116 |
| GDP.pdf | 6.0% | 2.4 min | 9628 |
| CritPt | 0.6% | 10.0 min | 39364 |
| AA-Omniscience | 76.7% | 0.6 min | 2386 |
| AA-LCR (long context) | 78.3% | 0.3 min | 1091 |
Run MiniMax 2.7 privately
One key, one endpoint, three jurisdictions.
- Size tierLarge
- Serving routeStandard
- Input rate$0.84/1M
- Output rate$2.10/1M
- Cached input rate—
- Context window180K tokens
- Max output16K tokens
- ReasoningYes — emits thinking traces
- Vision & video inputText only
- Tool callingSupported
- Data residencyAU, EU and US endpoints — processing stays inside the region you call