← the radar · ✳ model tracker
MiniMax M3.1 Flash
● live- Released
- October 19, 2026
- Before launch
- "Space Bunny" on OpenCode, 26 days
- Price
- $0.15 / $0.60 per 1M tokens
- Context
- 1M tokens
- Bench index
- 35.2 · #31 of 50
The watchers keep tracking it: lineup changes, pricing moves and deprecations all alert.
Launch details
Live — Oct 19; rolled out Oct 19 (MiniMax's launch post); $0.15/$0.60 per 1M, cached input $0.015, flat; 1M context; on the MiniMax API, M Plan and MiniMax Code; weights 'will be released later'; MiniMax's own figures: DeepSWE v1.1 73.7%, OSWorld V2 61.7, Agents' Last Exam 29.6, CyberGym 82%, BankerToolBench 76.2; MiniMax says agentic tasks cost 70% less than on its predecessor, with 200+ tokens/s output; Space Bunny, the anonymous stealth model on OpenRouter and OpenCode from Sep 23, was an early version, per MiniMax; Artificial Analysis Intelligence Index 42 on launch day, against 45 for DeepSeek V4.1 Flash 1017 at the same off-peak price; preview Sep 27 in MiniMax Code as M3.1-Flash-Preview.Answering in 0.7s· answered Live model status →
Watch for MiniMax’s next model, free →Launch report (PDF)
Timeline
MiniMax officially rolled out M3.1 Flash on October 19 and named it as the model behind Space Bunny, the anonymous stealth model that appeared free on OpenRouter and OpenCode on September 23.
On the bench index
Coding leaderboard
#37 of 47 on the coding index, at 30.1 · combines LMArena · Coding, LiveBench · Coding, Vals Vibe Code Bench
Open weights
#7 of 17 open-weights models on the Bench Index · MiniMaxAI/MiniMax-M3.1-Flash on Hugging Face · 220B parameters · license custom (minimax-community)
Where to use MiniMax M3.1 Flash
27 providers list MiniMax M3.1 Flash in the models.dev catalog (as of 2026-11-12), under the model ids shown below.
MiniMax-M3.1-FlashMiniMax-M3.1-Flashminimax/minimax-m3.1-flashminimax/minimax-m3.1-flashminimax-m3.1-flashminimax-m3.1-flashminimax-m3.1-flashminimax/MiniMax-M3.1-Flashminimax-m3-1-flashaccounts/fireworks/models/minimax-m3p1-flashMiniMaxAI/MiniMax-M3.1-Flashminimax/minimax-m3.1-flashAlso listed by LLM Gateway, Merge Gateway, MiniMax Token Plan (minimax.cn), MiniMax Token Plan (minimax.io), Modular, NanoGPT, Nebius Token Factory, Novita AI, Nvidia, Ollama Cloud, OpenCode, OpenCode Go, Requesty, Venice AI, ZenMux.
Listings cover the model family, so a provider may list a sibling tier under the same name. Check the id before you build on it.
MiniMax M3.1 Flash: quick answers
Is MiniMax M3.1 Flash out?
Yes. MiniMax M3.1 Flash is live and has been since October 19, 2026.
When was MiniMax M3.1 Flash released?
October 19, 2026.
Is MiniMax M3.1 Flash good for coding?
On the ArtificialWatch coding index it scores 30.1, #37 of 47 ranked models. The index combines LMArena · Coding, LiveBench · Coding and Vals Vibe Code Bench, taking the median of each model's percentile across them.
Is MiniMax M3.1 Flash open source?
It has open weights: the trained model is public on Hugging Face as MiniMaxAI/MiniMax-M3.1-Flash, 220B parameters as Hugging Face counts them, license custom (minimax-community). Among the 17 open-weights models the ArtificialWatch Bench Index ranks, it is #7.
Who makes MiniMax M3.1 Flash?
MiniMax. ArtificialWatch also tracks MiniMax-H3, MiniMax M3, MiniMax M3.1 Pro, MiniMax M3.5 and MiniMax M4 from MiniMax.
Where can I use MiniMax M3.1 Flash?
27 providers list it in the models.dev catalog, including MiniMax (minimax.cn), MiniMax (minimax.io), OpenRouter, Vercel AI Gateway and Cortecs. The exact model ids each one lists are on this page.
How much does MiniMax M3.1 Flash cost?
MiniMax's own API lists MiniMax M3.1 Flash (MiniMax-M3.1-Flash) at $0.15 per million input tokens and $0.60 per million output tokens, and $0.015 per million cached input tokens. Prices as published in the models.dev catalog, as of 2026-11-12.
What is MiniMax M3.1 Flash's context window?
1M tokens, with up to 524K output tokens per response, as MiniMax lists MiniMax M3.1 Flash (MiniMax-M3.1-Flash).
Was MiniMax M3.1 Flash tested anonymously before launch?
Yes. OpenCode listed it as the anonymous "Space Bunny" from September 23, 2026, and it was named on October 19, 2026, 26 days later: OpenCode revealed it as MiniMax M3.1 Flash Free, the day MiniMax launched the model. Every anonymous OpenCode model and the lab behind it is on ArtificialWatch's stealth models page.
Is MiniMax M3.1 Flash open weights?
Yes. Its weights are public on Hugging Face as MiniMaxAI/MiniMax-M3.1-Flash.
Every live model's API price, side by side → · What a workload costs on MiniMax models →
More from MiniMax
Cite this page
Writing about MiniMax M3.1 Flash? Copy a link or a citation.
<a href="https://artificialwatch.com/model/minimax-m3-1">MiniMax M3.1 Flash on ArtificialWatch</a>ArtificialWatch, "MiniMax M3.1 Flash", https://artificialwatch.com/model/minimax-m3-1Know when MiniMax's next model drops. Free.
Free: a Chrome push and an email about 15 minutes after it's confirmed on a public API. Members get a text or a phone call the second it lands.
- Free forever
- No card
- Unsubscribe in one click
Also: the wire · the bench index · our track record