Gemini 4 Hydrogen vs Claude Opus 5
Gemini 4 Hydrogen is from Google, released November 12, 2026. Claude Opus 5 is from Anthropic, released July 24, 2026.
- LMArena Elo1490 vs 1490. Both rated 1490 on LMArena.
- Input per 1M$1.50 vs $5. Gemini 4 Hydrogen costs 70% less per million input tokens.
- Output per 1M$7.50 vs $25. Gemini 4 Hydrogen costs 70% less per million output tokens.
- Context window1.05M vs 1M. Gemini 4 Hydrogen takes 4.9% more context in one request.
Prices are the vendors’ published API rates; Elo is LMArena’s text board. Each figure is dated on its model page.
Interactive Comparison
Loading chart...
Token cost calculator
Input tokens
Output tokens
| Model | Input | Output | Cost for this many tokens |
|---|---|---|---|
| Gemini 4 Hydrogen | $0.15 | $0.15 | $0.30 |
| Claude Opus 5 | $0.50 | $0.50 | $1.00 |
List prices per 1M tokens from each model's record (Gemini 4 Hydrogen: $1.5 in / $7.5 out; Claude Opus 5: $5 in / $25 out). Cached-input and batch discounts are not included. Click a figure to type it. Compare up to 8 models in the token cost calculator.
$1.50 / $7.50 per million tokens | $5 / $25 per million tokens | |
|---|---|---|
| Identity | ||
| Developer | Anthropic | |
| Released | Nov 2026 | Jul 2026 |
| Status | GA - generally available in the Gemini API since November 12, 2026 | Available (legacy) |
| Licence | Proprietary | Proprietary |
| Self-hostable | No | No |
| Cost | ||
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $3.000 / 1M tokens | $10.000 / 1M tokens |
| Input price | $1.500 / 1M tokens | $5.000 / 1M tokens |
| Output price | $7.500 / 1M tokens | $25.000 / 1M tokens |
| Cached input | $0.075 per 1M tokens (implicit caching only) | $0.50 / 1M tokens (cache hit) |
| Batch discount | — | 50% off input/output (Batch API) |
| Free tier | Yes - Gemini API free tier and Google AI Studio (rate-limited) | Small trial credit for new accounts, no ongoing free tier |
| Capacity | ||
| Context window | 1,048,576 tokens | 1M tokens |
| Max output | 65,536 tokens by default; up to 1,048,576 with long decode | 128k tokens |
| Long-context surcharge | — | None (full 1M window at standard pricing) |
| Capability | ||
| Vision in | Yes | Yes |
| Audio in | Yes | No |
| Function calling | Yes | Yes |
| Structured output | Yes | Yes |
| Extended reasoning | Yes - thinking levels low, medium, high (default) and xhigh; minimal is rejected | Yes |
| Web search | Yes - Grounding with Google Search and Google Maps | Yes |
| Code execution | Yes | Yes |
| Access | ||
| API | Yes - Gemini API (GenerateContent and Interactions API), free and paid tiers | Yes |
| Chat app | Yes - Gemini app and Google Search; Google Antigravity for paid Google AI Pro and Ultra subscribers | Yes |
| Cloud marketplaces | — | AWS Bedrock, Google Cloud, Microsoft Foundry |
| Fine-tuning | — | No |
| Measured quality | ||
| LMArena Elo | 1490 (checked Nov 2026) | 1490 (checked Nov 2026) |
| LMArena rank | Rank 15 | Rank 16 |
| GPQA Diamond | — | 93 |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 497 | 149 |
| LMArena by task | ||
| Coding | 1518 Elo (rank 43) | 1532 Elo (rank 20) |
| Math | 1495 Elo (rank 21) | 1521 Elo (rank 3) |
| Creative Writing | 1492 Elo (rank 6) | 1471 Elo (rank 20) |
| Instruction Following | 1503 Elo (rank 6) | 1496 Elo (rank 11) |
| Hard Prompts | 1514 Elo (rank 15) | 1515 Elo (rank 14) |
| Longer Query | 1514 Elo (rank 8) | 1502 Elo (rank 14) |
| Multi-Turn | 1476 Elo (rank 50) | 1486 Elo (rank 28) |
| Expert | 1542 Elo (rank 6) | 1536 Elo (rank 7) |
