Satire. Every score on this site is made up. The tweets are real.Read the disclaimer →
Trust Me Bro
Get VerifiedContribute
ModelsBenchmarksIncidentsIn the WildMethodologyChart Crime Studio
TMB-2026-0014··Marketing claim
🤝🤝Partially delivered

Space Bunny, and the 93%

Subject: MiniMax M3.1 Flash

Claim

"Frontier-level intelligence", revealed to be the anonymous "Space Bunny" that "has topped global leaderboards with 200T+ tokens". Open weights promised for 23 October.

Finding

An independent index scored it 42, thirteen points above its predecessor: a real improvement. The same evaluator measured a 93% hallucination rate, meaning that when it does not know an answer it supplies one anyway. The weights shipped on 23 October, as promised. Severity is low because the bunny and the weights were both real.

Timeline

  1. 19 Oct 2026
    Space Bunny unmasked; "frontier-level intelligence".@MiniMax_AI · ♥ 7.5K ↗
  2. 19 Oct 2026
    Independent index: 42, level with GLM-5.3-Flash.@ArtificialAnlys · ♥ 702 ↗
  3. 20 Oct 2026
    93% hallucination rate: "when it misses, it rarely admits it."@mahmoudfettal · ♥ 1 ↗
  4. 23 Oct 2026
    Weights public, on the promised date.@MiniMax_AI · ♥ 2.4K ↗

Primary sources

The posts this entry is based on, embedded as published.

MiniMax_AI
MiniMax (official)
@MiniMax_AI
Link to tweet
MiniMax M3.1 Flash: Space Performance + Bunny Speed 🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash. 💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence
MiniMax M3.1 Flash: Space Performance + Bunny Speed 🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash. 💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence https://t.co/j0tCfNYYgo
♥ 7.5KOpen on X ↗
ArtificialAnlys
Artificial Analysis
@ArtificialAnlys
Link to tweet
MiniMax-M3.1-Flash scores 42 on the Artificial Analysis Intelligence Index, 13 points higher than MiniMax-M3 and less than half the cost per task. It scores level with GLM-5.3-Flash but behind DeepSeek V4.1 Flash 1017 MiniMax-M3.1-Flash is @MiniMax_AI's new Flash reasoning
MiniMax-M3.1-Flash scores 42 on the Artificial Analysis Intelligence Index, 13 points higher than MiniMax-M3 and less than half the cost per task. It scores level with GLM-5.3-Flash but behind DeepSeek V4.1 Flash 1017 MiniMax-M3.1-Flash is @MiniMax_AI's new Flash reasoning https://t.co/bEelNegMKP
♥ 705Open on X ↗
mahmoudfettal
Mahmoud 👨🏽‍💻🧙...
@mahmoudfettal
Link to tweet
MiniMax M3.1 Flash is free on OpenRouter, but Artificial Analysis found a 93% hallucination rate: when it misses, it rarely admits it. Also: Microsoft treats AI agents as insider threats, Amazon's TorchNeuron for Trainium. thedailyreboot.tech/episodes/daily… #AI #PyTorch
♥ 1Open on X ↗
MiniMax_AI
MiniMax (official)
@MiniMax_AI
Link to tweet
MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token. Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and
♥ 3.9KOpen on X ↗