MiniMax-M3.1-Flash ships: API first, weights later, 3x faster online at 70% lower cost
MiniMax shipped M3.1-Flash on Oct 19, the new Flash-series flagship, API first. Official claims: roughly 3x faster online inference, 70% lower task cost ($0.15 / $0.60 / $0.015 per 1M tokens), 200+ tok/s output. It ran anonymously as Space Bunny, processing 200 trillion tokens. Weights come later — logged as RESTRICTED with unpublished params, zero on-site runs.
Source: araoai
MiniMax-M3.1-Flash ships: API first, weights later, 3x faster online at 70% lower cost
On Oct 19, MiniMax shipped M3.1-Flash, the new Flash-series flagship, following an API-first, weights-later path: callable today, local runs wait on the official weights release.
Speed and price (official numbers)
| Metric | M3.1-Flash | vs predecessor |
|---|---|---|
| Online inference speed | ~3x | ~3x |
| Task total cost | — | down 70% |
| Output throughput | 200+ tok/s (P50 170+) | — |
| Input price | $0.15 / 1M tokens | — |
| Output price | $0.60 / 1M tokens | — |
| Cached input | $0.015 / 1M tokens | — |
More aggressive sparsity drives the speedup: the same tasks finish with fewer active parameters, and the official pitch pairs that with a 70% cut in task cost — for high-frequency agentic workloads that matters more than a benchmark delta.
Benchmarks (official numbers)
| Benchmark | Score |
|---|---|
| BankerToolBench | 76.2 |
| DeepSWE v1.1 | 73.7% |
| OSWorld V2 | 61.7 |
| Agents Last Exam | 29.6 (first among Chinese models on both leaderboards) |
| CyberGym | 82% |
First place overall and among Chinese models on Agents Last Exam, plus 82% on CyberGym, show the sparsity did not visibly cost agentic long-tail capability.
Space Bunny: run first, reveal later
Before launch the model ran anonymously as Space Bunny in open beta, processing 200 trillion tokens — the lab validated on real traffic before revealing the name. The run-first-reveal-later pattern is in vogue; for users it means the API stability was already load-tested before the banner went up.
How we logged it
Our catalog now lists MiniMax-M3.1-Flash (id 91): parameter count unpublished → params_b = NULL, weights not out → RESTRICTED. Same API-only stance as Kimi-K3 — zero on-site records; local speed and VRAM accounting wait for the weights.
Flash-field coordinates (existing catalog rows):
| Model | Total / active | Note |
|---|---|---|
| MiniMax-M3 | 427B | predecessor flagship |
| GLM-5.3-Flash | 320B / A32B | Zhipu Flash tier |
| Qwen3.8-Flash-Next | 180B / A14B | Alibaba Flash tier |
Availability
Available now through the MiniMax API, M Plan and MiniMax Code. Local deployers wait on the weights; on MiniMax's recent cadence the weights window has usually opened within weeks of the API launch.