ACB
70.2 score
LLM Leaderboard · Proprietary
Gemini 4 Hydrogen benchmarks — leaderboard rank, pricing, and comparisons.
Gemini 4 Hydrogen ranks #13 on the AskClash LLM leaderboard with an overall score of 62.8. It scores 54.0 on Terminal-Bench 4.0. Compare Gemini 4 Hydrogen vs GPT, Claude, Gemini, DeepSeek, open-weight, and frontier AI models using public benchmark scores, token pricing, context window, and access details.
Rank #13AskClash overall score: 62.8
$1.50 / $7.50Input and output token price, when published. Context: 1M.
Visit websiteVisit the model provider's website.
Gemini 4 Hydrogen benchmark snapshot
AskClash combines public LLM benchmark cells into a weighted percentile score and penalizes missing coverage so narrow rows do not dominate better-measured models.
Overall62.8
Benchmark cells5
Context1M
CreatorGoogle
Gemini 4 Hydrogen public benchmark scores
Cached benchmark values can include HLE, GPQA, SWE-bench, SWE-Pro, SWE-Atlas, Terminal-Bench, MCP Atlas, MMMU-Pro, ARC-AGI-2, Tau2, and model-specific coding or agent scores.
AA Intelligence Index
53.9 score
AutomationBench-AA
76.1 score
Terminal-Bench 4.0
54.0 score
Finance Agent v2.1
65.1 score
Gemini 4 Hydrogen: its diagnostic ACB uses a review slice and sampled video frames.
More published benchmark results
Versioned source-specific results; Google self-evaluations and independent evaluator results are distinct.
Google model card
- SWE-sweep: 13.5%
- Terminal-Bench 4.1 (cheating-adjusted): 57.0%
- Agents' Last Exam v1.1: 71.6%
- AutomationBench-AA: 76.1%
- Vending-Bench 2 balance: $14,549
- OSWorld-2.1 offline subset: 77.6%
- MKA Climb WSR: 53.4%
- Chartography: 66.6%
- Terminal-Bench-Science 0.1: 51.4%
- HLE-Diamond: 58.8%
Artificial Analysis xhigh
- HLE-Diamond: 58.8%
- FrontierCode-AA 2.0: 38.96%
- Terminal-Bench-Science 0.1: 49.52%
- GDP.pdf all-pass (xHigh): 27.6%
Artificial Analysis medium
- GDP.pdf all-pass (Medium): 32.8%
Cursor
- CursorBench 4.0 (High): 54.3%
- CursorBench 4.0 (Medium): 46.7%
- CursorBench 4.0 (Low): 27.9%
Cognition Devin
- FrontierCode 2.0: 38.2%
Mercor
- APEX-Agents 1.1 Pass@1: 78.0%
- APEX-Agents Corporate Law Pass@1: 73.8%
- APEX-Agents Management Consulting Pass@1: 88.4%
- APEX-Agents Investment Banking Pass@1: 71.9%
- APEX-Accounting Pass@1: 20.3%
Vals AI
- Vals Index v2.2: 68.49%
- EMB v1.1: 75.80%
- Legal Research Bench: 53.85%
- Tax Agent Bench: 52.30%
- Vibe Code Bench v1.1: 87.78%
- Terminal-Bench 4.0 (Vals): 54.04%
Gemini 4 Hydrogen vs other AI models
Use these comparison links to evaluate Gemini 4 Hydrogen against nearby LLMs by benchmark score, price, context window, and provider.
Related AI and tech coverage
Cached AskClash article matches that can provide release, provider, benchmark, pricing, or market context around this model.
Artificial Analysis' Intelligence Index: Gemini 4 Hydrogen nearly ties Claude Sonnet 5.5 and GPT-6.1 Sol, at a third of Sonnet's cost but ~2.5x that of Sol (@artificialanlys)
@artificialanlys : Artificial Analysis' Intelligence Index: Gemini 4 Hydrogen nearly ties Claude Sonnet 5.5 and GPT-6.1 Sol, at a third of Sonnet's cost but ~2.5x that of Sol — Google's new Gemini 4 Hydrogen has Intelligence just below Claude Sonnet 5.5 and GP
Google's Gemini 4 Hydrogen outscores the bigger Argon just four weeks after its launch
Google's Gemini 4 Hydrogen replaces the Flash series, but benchmarks suggest it also makes the pricier Argon obsolete. On the independent Artificial Analysis Intelligence Index, Hydrogen nearly matches Claude Sonnet 5.5 for less money but costs about 2.5 times
Google rolls out Gemini 4 Hydrogen, finally replacing 3.8 Flash
Ars Technica
Google says Gemini 4 Hydrogen is available across AI Studio, the Gemini API, Search, paid AI Pro and Ultra tiers in Antigravity, and more (Abner Li/9to5Google)
Abner Li / 9to5Google : Google says Gemini 4 Hydrogen is available across AI Studio, the Gemini API, Search, paid AI Pro and Ultra tiers in Antigravity, and more — After this week's Carbon availability, Google today unveiled Gemini 4 Hydrogen instead of keepin
Last cached leaderboard date: . This model page is generated from the AskClash LLM Leaderboard cache and linked from the live leaderboard.