AIBreaking Wire
Pricing

AI Model Benchmarks

Track AI model performance across industry-standard benchmarks. Data sourced from official provider reports and third-party evaluations.

MMLU

Massive Multitask Language Understanding — general knowledge across 57 subjects. 2026 frontier values are estimates (providers no longer publish them).

HumanEval

Code generation benchmark — function-level completion from docstrings. 2026 frontier values are estimates.

GPQA

Graduate-level Google-Proof Q&A — expert-written science questions (Diamond subset)

SWE-bench

Real-world software engineering tasks from GitHub issues (Verified subset)

Score by Benchmark

Full Comparison Table

Sorted by average score. ★ = top score in column.

ModelProviderMMLUHumanEvalGPQASWE-benchDate
#1Claude Fable 5SuspendedAnthropic★93.0★96.092.6★95.02026-06
Claude Opus 4.8Anthropic92.591.091.090.02026-05
GPT-5.5OpenAI92.591.093.582.62026-04
Gemini 3.1 ProGoogle91.085.0★94.380.62026-02
Muse SparkMeta88.080.089.577.42026-04
Grok 4.3xAI87.580.090.172.02026-04
Claude Sonnet 4.6Anthropic89.084.080.072.02026-03
DeepSeek-V3.2DeepSeek85.075.082.473.12025-12
GPT-4oOpenAI87.278.655.148.02024-05

Scores represent pass@1 or accuracy percentages as reported by providers or third-party evaluations. The dataset is refreshed weekly by an agent that web-searches provider announcements and leaderboards. GPQA (Diamond) and SWE-bench (Verified) figures are sourced; most 2026 frontier labs no longer publish classic MMLU / HumanEval headline numbers, so those columns for current models are directional estimates rather than provider claims. Models flagged “Suspended” may be temporarily unavailable — verify current status with the provider.

AI Breaking Wire

The pulse of artificial intelligence — breaking news, security, tools, and platform tracking, refreshed every four hours by an AI newsroom.

Last build · 2026-07-17

The AI Brief

Free weekly digest — top AI news, tools, and security alerts.

Explore

  • News
  • Tools
  • Jobs
  • Merch
  • Webinars
  • Dashboards

Community

  • Discord
  • Projects
  • Marketplace
  • Claude Code
  • Events

Security

  • Security Hub
  • Vulnerability DB
  • Security News
  • Challenges

Company

  • About
  • Live Edition
  • Editorial Desks
  • Your Feed
  • Contact
  • Pricing
  • Advertise
  • Forge Portal
  • Editorial Policy
  • Privacy
  • Terms

Developers

  • API Docs
  • API Keys

Connect

  • Discord
  • Twitter / X
  • GitHub
  • Newsletter
  • Newsletter Archive
  • RSS Feeds

© 2026 AI Breaking Wire · Editorial standards uphold accuracy and AI transparency · See Editorial Policy and Privacy.

Press tip line: [email protected]

Massive Multitask Language Understanding — general knowledge across 57 subjects. 2026 frontier values are estimates (providers no longer publish them).

Anthropic
OpenAI
Google
Meta
xAI
Mistral
DeepSeek