Gemini 4 Hydrogen scores 54 on AA Index, 1 point behind Sonnet 5.5 at ~32% of its cost
ArtificialAnlys · x · 2026-11-13
Artificial Analysis reviews Gemini 4 Hydrogen, Google DeepMind's third new model in six weeks, which debuts an xhigh reasoning tier.
- Scores 54 on the Intelligence Index at xhigh — 1 point behind Claude Sonnet 5.5 (max) and GPT-6.1 Sol (max), 4 ahead of Gemini 4 Argon (high); 51 at default high
- Priced $1.50/$7.50 per M tokens; xhigh costs 32% as much per task as Sonnet 5.5 for near-equal intelligence, but 2.5x GPT-6.1 Sol; high drops Cost per Task to $1.87
- Strengths: visual reasoning (33% all-pass on GDP.pdf, tied #1), 8% hallucination rate vs 43% (Sonnet 5.5) and 52% (Sol), strong presentation on AA-Briefcase
- Weaknesses: Terminal-Bench 4.0 at 54% and Terminal-Bench-Science at 50%, both behind Sonnet 5.5 (64%/53%); xhigh nearly doubles output tokens for only a 3-point gain
- 200 output tokens/sec, sits on the Time-per-Task vs Intelligence Pareto frontier at high/medium/low; 1M context window
Related event: Gemini 4 Hydrogen Nears Sonnet 5.5 Intelligence at a Third of the Cost(4 posts)→
More from Models
- Pro user vents at OpenAI's long wait: 'GPT-6.2 better be good' — Angaisb_ · 2026-11-13
- Brave quietly publishes Leo browser assistant's custom VL models and datasets on Hugging Face — gnukeith · 2026-11-13
- Gemini 4 Hydrogen takes top spot on FinAgent Arena by a wide margin — osanseviero · 2026-11-13
- Thorsten-4 model reads execs' false confidence on earnings calls; hedge fund trades $50M on it — rohanpaul_ai · 2026-11-13
- Researcher blasts Gemini's confusing naming: just tell us which model is bigger — DimitrisPapail · 2026-11-13
- Leaked Gemini 4 Hydrogen benchmarks claim agent wins over Sonnet 5.5 at $1.50/M input tokens — heypearlai · 2026-11-13