LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS

Leaderboards

Data from Artificial Analysis · Intelligence Index v5 · 195 models with data

Value frontier →

Artificial Analysis Intelligence Index. Score uses the “representative tier” (highest-scoring reasoning tier per model).

Leaders of every board

The top 8 on each board, with the metric that ranks them. Open any model for its full pricing across providers.

Intelligence

  • Claude Fable 5.2AA 64.6
  • GPT-6.2 AstraAA 62.1
  • Gemini 4 NeonAA 61.4
  • Gemini 4 CarbonAA 60
  • Claude Opus 5.5AA 59.1
  • GPT-6.2 SolAA 57.3
  • GPT-6 AstraAA 56.9
  • GPT-6.1 SolAA 54.6

Engineering

  • Claude Fable 5.272.6
  • Gemini 4 Neon66.8
  • GPT-6.2 Astra66.6
  • Claude Opus 5.565.3
  • Gemini 4 Carbon65.3
  • GPT-6 Astra61.8
  • GPT-6.2 Sol61.2
  • Gemini 4 Hydrogen60.1

Value

  • Muse Spark 1.3293
  • Claude Haiku 5.5183
  • Muse Ember180
  • DeepSeek V4.1 Flash 1017124
  • Ling 3.1 Flash66
  • Inkling 2 Small64
  • MiMo-V2.6-Pro60
  • GLM-5.329

Cost per task

  • Mistral Small 4$0.04
  • Granite 4.2 8B$0.041
  • MiMo-V2.6-Flash$0.13
  • Muse Ember$0.14
  • GPT-6 Luna$0.15
  • LFM3 1.7B$0.17
  • Gemini 3.1 Flash Lite$0.19
  • Inkling Small$0.20

Speed

  • Celeris 11543 tok/s
  • Mercury 2.5793 tok/s
  • Muse Ember659 tok/s
  • Mercury 2563 tok/s
  • Sarvam 105B413 tok/s
  • Gemini 3.5 Flash Lite402 tok/s
  • Gemini 3.1 Flash Lite322 tok/s
  • Nemotron 3 Nano 30B A3B318 tok/s

Usage

  • DeepSeek V4.1 Flash 1017312.0T
  • Hy4 flash69.3T
  • GLM-5.3-Flash56.7T
  • GPT-6 Luna36.8T
  • Hy435.9T
  • Claude Opus 5.531.2T
  • MiniMax-M3.1-Flash26.4T
  • Step 5 Preview25.9T

GPQA Diamond

  • GPT-6 Astra96.1%
  • Gemini 3.8 Flash95.3%
  • Gemini 3.1 Pro Preview94.1%
  • Qwen3.8 2.4T A95B93.5%
  • MiniMax-M392.9%
  • DeepSeek V4 Pro 081392.8%
  • Qwen3.8 Max92.8%
  • Qwen3.8 Flash Next92.3%

Reading the leaderboards

Each board ranks the same catalog on a different axis: raw intelligence, Artificial Analysis domain indices (engineering, finance, legal, healthcare, strategy and economics), output speed, cost per task, real 30-day usage, and value. The value board divides intelligence by blended price, with an intelligence floor so a cheap but weak model cannot top it on price alone.

The price basis is switchable because the lowest channel is almost always a third-party gateway, while the official list price reflects the first-party rate. Cost per task can reorder the ranking sharply: a reasoning model with a low unit price still runs up a high bill when it emits many thinking tokens per answer.