Updated daily at 05:00 UTC
Vision-language
Gemini Robotics
Gemini embodied-reasoning models for spatial understanding and task planning in robots.
By Google DeepMind.
Versions
4 versions
| Model | Maker | Kind | Released | Context | Input $/1M | Output $/1M |
|---|---|---|---|---|---|---|
| Gemini Robotics-ER 2Gemini Robotics | Google DeepMind | Vision-language | 30 Jul 2026 | 131.1K | $1.00 | $5.00 |
| Gemini Robotics-ER 2 StreamingGemini Robotics | Google DeepMind | Vision-language | 30 Jul 2026 | 131.1K | $1.00 | $5.00 |
| Gemini Robotics-ER 1.6Gemini Robotics · Retired | Google DeepMind | Vision-language | 14 Apr 2026 | 131.1K | $1.00 | $5.00 |
| Gemini Robotics-ER 1.5Gemini Robotics · Retired | Google DeepMind | Vision-language | 25 Sep 2025 | 1M | $0.30 | $2.50 |
- Gemini Robotics-ER 2Google DeepMind · 30 Jul 2026 · 131.1K context$1.00in / 1M
- Gemini Robotics-ER 2 StreamingGoogle DeepMind · 30 Jul 2026 · 131.1K context$1.00in / 1M
- Gemini Robotics-ER 1.6Google DeepMind · 14 Apr 2026 · 131.1K context$1.00in / 1M
- Gemini Robotics-ER 1.5Google DeepMind · 25 Sep 2025 · 1M context$0.30in / 1M