Pyyan / Compare / Gemini 3.8 Live Extended Thinking vs GPT-6 Astra
Gemini 3.8 Live Extended Thinking vs GPT-6 Astra
2 of 5Language Models · verified 16 Sept 2026
Gemini 3.8 Live Extended ThinkingGooglecurrent
GPT-6 AstraOpenAIcurrent| Specification | Gemini 3.8 Live Extended Thinking | GPT-6 Astra |
|---|---|---|
| Summary | Top of the speech to speech board, and it reasons out loud while it gets there. | Saturates three of the benchmarks it reports, and ships disabled for enterprise. |
| Context | Not published | 1.05M |
| Input ($/Mtok) | $0.75 | $10 |
| Output ($/Mtok) | $4.5 | $50 |
| Licence | Proprietary | Proprietary |
| Max output | Not published | 128K |
| Parameters | Not published | Not published |
| Released | 15 Sep 2026 | 3 Sep 2026 |
| Model ID | gemini-3.8-live-extended-thinking | gpt-6-astra |
| Category | Language Models | Language Models |
| Official | Google ↗ | — |
Highlighted rows are where these differ.
Gemini 3.8 Live Extended Thinking
- First on the Artificial Analysis Speech to Speech Quality Index at 82.6
- 68.6% on tau-Voice and 97.7% on Big Bench Audio, both reported by Google
- Speaks while it reasons, with cues such as let me check that
Best for hard tasks done by voice.
GPT-6 Astra
- 97.6% on FrontierMath Tier 4, 99.9% on ARC-AGI-3 and 100% on ExploitBench
- Above 272K input tokens the whole request reprices: input and cache at 2x, output at 1.5x
- Rolls out to OpenAI's Daybreak cyber programme first, and arrives switched off for enterprise
Best for hard reasoning and computer use.