Gemini 4 Hydrogen: a fast robotics model now in the Gemini API
Google released Gemini 4 Hydrogen, a fast model for robotics, with the model ID gemini-4-hydrogen.
Models:Gemini Robotics ER 2
On Zabio today
gemini-4-hydrogenLive. Call it by name, or let auto pick it when it fits the job.Price and detailsgemini-4-argonLive. Call it by name, or let auto pick it when it fits the job.Price and detailsgemini-4-carbonLive. Call it by name, or let auto pick it when it fits the job.Price and details
What changed
Google released Gemini 4 Hydrogen, a fast and efficient model built for robotics, with strong spatial and embodied reasoning. Developers can use it in the Gemini API through the Interactions API with the model ID gemini-4-hydrogen. Hydrogen scores 58 on the Artificial Analysis Robotics 0.1 benchmark, similar to Argon at a lower cost per token. Google also launched the Robotics plugin for Google Antigravity on the same day, which adds skills for hardware onboarding, MuJoCo simulation and robot control.
Key details
- Model ID: gemini-4-hydrogen.
- Day-one support in the google-genai SDKs.
- Score of 58 on the Artificial Analysis Robotics 0.1 benchmark.
- On the public ERQA benchmark, Hydrogen outperforms Gemini Robotics ER 2 on spatial reasoning tasks such as multi-view correspondence and 3D trajectory reasoning.
- Google AI Studio lets you try spatial prompts, live camera streams, and structured tool calls in the browser.
- Google Antigravity is for paid Google AI Pro or Ultra subscribers.
What it means for Zabio customers
Google recommends that developers using Gemini Robotics ER 2 for these workloads try Gemini 4 Hydrogen.
Source
Google’s Gemini 4 Hydrogen brings strong spatial and embodied reasoning capabilities to a fast, efficient model built specifically for robotics.
Original announcement: Google for Developers blog
How it compares
The models in this story next to models of the same class at a similar price, all available with one Zabio key.
| Model | Input, $/M | Output, $/M | Cached, $/M | Context window | 1,000 calls | Writing code, hard |
|---|---|---|---|---|---|---|
gemini-4-hydrogenGoogleIn this story | 1.50 | 7.50 | no data | 200,000 | $6.75 | 61.8 |
gemini-4-argonGoogleIn this story | 2.00 | 10.00 | no data | 200,000 | $9.00 | 58.7 |
gemini-4-carbonGoogleIn this story | 2.00 | 10.00 | no data | 200,000 | $9.00 | 72.7 |
qwen3.5-omni-plusQwen | 1.40 | 8.30 | no data | 196,608 | $6.95 | no data |
gpt-5OpenAI | 1.25 | 10.00 | 0.125 | 400,000 | $7.50 | 4.4 |
claude-sonnet-5Anthropic | 2.00 | 10.00 | 0.20 | 1,000,000 | $9.00 | 63.3 |
Prices are the vendor list prices in USD per million tokens. "1,000 calls" is 1,000 requests of 2,000 input tokens and 500 output tokens each, without caching. Zabio bills the list price plus its service fee: see pricing.
Scores: percent correct for that kind of job, aggregated from public results by Epoch AI, LMArena, and LiveBench, read on 16 November 2026. Epoch AI: CC BY 4.0, LMArena: CC BY 4.0, LiveBench: Leaderboard results used with attribution; the 2.0 release states no terms of its own (read 2026-11-06); the first version's Mathematics, Language, IF results: Code Apache 2.0 and MIT; leaderboard results used with attribution
Source: Google for Developers blog (2026-11-17)