TPSTokens Per Second
NewsModelsIDEsRankingsBenchmarksSecurity

Browse

NewsModels└Open Source & LocalIDEsChangelog

Compare

RankingsBenchmarksModel QuizSecurity

Yours

FollowingSavedHistoryWhat I Missed

Companies

AnthropicOpenAIDeepMindMeta AIDeepSeekxAIMistral AIPerplexity AIAll companies →

MiniMax

Chinese AI company, maker of MiniMax and Hailuo models

https://www.minimaxi.com →

Benchmark Highlights (models from the last 12 months)

Full leaderboard →

Arena Elo

1445 Elo

MiniMax-M3

SWE-bench Verified

80.5%

MiniMax-M3

GPQA

92.9%

MiniMax-M3

LiveCodeBench

82.6%

MiniMax-M2

MMLU-Pro

82%

MiniMax-M2

MMMU

78.1%

MiniMax-M3

Hallucination Rate (lower is better)

18.4%

MiniMax-M3

Speed (tok/s)

114 tok/s

MiniMax-M2

News

No articles yet.

Latest Models

All 9 models →

MiniMax H3

MiniMax

Open weights
Context—
In /1M—
Out /1M—

Released August 3, 2026

MiniMax-M3

MiniMax

Open weights
Context1.0M
In /1M$0.3
Out /1M$1.2

Released June 2, 2026

MiniMax M2.7

MiniMax

Context205K
In /1M$0.3
Out /1M$1.2

Released March 16, 2026

MiniMax-M2.5

MiniMax

Open weights
Context197K
In /1M$0.3
Out /1M$1.1

Released February 12, 2026

MiniMax-M2

MiniMax

Open weights
Context1.0M
In /1M$0.3
Out /1M$1.2

Released October 26, 2025

MiniMax-M1

MiniMax

Open weights
Context1.0M
In /1M$0.4
Out /1M$2.2

Released June 17, 2025

MiniMax-Text-01

MiniMax

Open weights
Context1.0M
In /1M$0.2
Out /1M$1.1

Released January 15, 2025

TPS

Tokens Per Second. The fastest LLM news on the internet — tracked automatically every 15 minutes.

Coverage

  • Latest News
  • Model Database
  • AI IDEs
  • Compare IDEs
  • Changelog
  • Benchmarks
  • Compare Models

Companies

  • Anthropic
  • OpenAI
  • Google DeepMind
  • Meta AI
  • DeepSeek
  • xAI
  • Mistral AI
  • Perplexity AI
  • All companies →

Guides

  • All Rankings
  • Best by Use Case
  • Best Coding LLM
  • Best Cheap LLM
  • Best Reasoning LLM
  • Best AI IDE
  • Compare Models
  • Compare IDEs
  • About TPS
  • RSS Feed
  • Atom Feed

© 2026 TPS — Tokens Per Second.

The fastest LLM news. All signal, no noise.