Gemini 4 Hydrogen adds xhigh thinking level, with official guidance on all four tiers
patloeber · x · 2026-11-17
Google's Gemini 4 Hydrogen uses a thinking level as the main control over quality, latency, and token use, and introduces a new xhigh tier:
- low: fewest tokens, lowest latency — for cost/latency-sensitive workloads like fast chat and high-throughput processing; still capable on agentic tasks
- medium: faster responses — for interactive coding, live UI prototyping, and latency-sensitive tool calls
- high (default): balanced intelligence and speed — recommended for most tasks including complex coding, agentic execution, tool calling, and multimodal knowledge work
- xhigh (new): deepest reasoning, most tokens — reserve for hard multi-step problems where high falls short; can cause overthinking on simple prompts
Official advice: start with high, run your evals at all four levels with a high maxoutputtokens so answers aren't cut short, then pick the lowest thinking level that meets your quality bar.
More from coding & agent
- REA-Video skill lets Claude reverse-engineer video editing styles; rival claims 20x speed at SSIM 0.96 — toolstelegraph · 2026-11-17
- Developer retries Amp Code after months and finally gets the hype: "completely different" — g3off_who · 2026-11-17
- Dev shares 5 AI automations for service businesses that actually make owners money — Crazy-Movie7633 · 2026-11-17
- SNEWPapers launches remote MCP turning AI tools into US newspaper archive researchers — Jolly-Newspaper-6769 · 2026-11-17
- Team designing open-source mock MCP server, asks community for testing pain points first — OkSkirt1039 · 2026-11-17
- Dev publishes live agent key with ¥100/day cap and challenges the internet to break it — kokesi12 · 2026-11-17