Claude Haiku 5.5 vs Gemini 3.8 Flash
Claude Haiku 5.5 is from Anthropic. Gemini 3.8 Flash is from Google. Claude Haiku 5.5 costs $0.500 per million output tokens. Gemini 3.8 Flash costs $4.13. Claude Haiku 5.5 leads on the Epoch Capabilities Index: Claude Haiku 5.5 scores 160.4 and Gemini 3.8 Flash scores 157.6. A typical request, about 1,000 words in and 400 words out, costs $0.0004 on Claude Haiku 5.5 and $0.0031 on Gemini 3.8 Flash.
Prices updated . Benchmarks updated .
| Claude Haiku 5.5 | Gemini 3.8 Flash | |
|---|---|---|
| Lab | Anthropic | |
| You send / 1M | $0.100 | $0.825 |
| It replies / 1M | $0.500 | $4.13 |
| Context | 1M | 1.05M |
| Launched | 2026-10-07 | 2026-09-02 |
| Weights | Closed | Closed |
| Typical request | $0.0004 | $0.0031 |
| ECI | 160.4 | 157.6 |
| GPQA | 89.7% | 95.4% |
| ARC-AGI-2 | 61.3% | 89.2% |
| HLE | 30.1% | 44.5% |
| AIME | 98.9% | 98.9% |
| SimpleQA | 23.8% | 69.7% |
| CritPt | 22.6% | 18.3% |
| Arena | 1,401 | 1,498 |
| Arena web | 1,589 | 1,583 |
Which is cheaper?
Claude Haiku 5.5 costs less to run: $0.500 per million output tokens, against $4.13 for Gemini 3.8 Flash. Input tokens are $0.100 per million for Claude Haiku 5.5 and $0.825 per million for Gemini 3.8 Flash. A typical request, about 1,000 words in and 400 words out, costs $0.0004 on Claude Haiku 5.5 and $0.0031 on Gemini 3.8 Flash.
Which scores higher?
Claude Haiku 5.5 scores higher on Epoch Capabilities Index: 160.4 against 157.6 for Gemini 3.8 Flash. That index is an independent composite from Epoch AI. Lab-reported scores are kept off this comparison. Context windows are 1M and 1.05M. Prices are US dollars per million tokens at current public rates.
Verdict
Claude Haiku 5.5 is both cheaper and higher on Epoch Capabilities Index than Gemini 3.8 Flash, at $0.500 per million output tokens and a score of 160.4. Context is 1M for Claude Haiku 5.5 and 1.05M for Gemini 3.8 Flash. Use the model pages for every score on file, including lab-reported numbers.
FAQ
Which is cheaper, Claude Haiku 5.5 vs Gemini 3.8 Flash?
Claude Haiku 5.5 costs less to run: $0.500 per million output tokens, against $4.13 for Gemini 3.8 Flash. Input tokens are $0.100 per million for Claude Haiku 5.5 and $0.825 per million for Gemini 3.8 Flash. A typical request, about 1,000 words in and 400 words out, costs $0.0004 on Claude Haiku 5.5 and $0.0031 on Gemini 3.8 Flash.
Which scores higher, Claude Haiku 5.5 vs Gemini 3.8 Flash?
Claude Haiku 5.5 scores higher on Epoch Capabilities Index: 160.4 against 157.6 for Gemini 3.8 Flash. That index is an independent composite from Epoch AI. Lab-reported scores are kept off this comparison. Context windows are 1M and 1.05M. Prices are US dollars per million tokens at current public rates.
What is the verdict for Claude Haiku 5.5 vs Gemini 3.8 Flash?
Claude Haiku 5.5 is both cheaper and higher on Epoch Capabilities Index than Gemini 3.8 Flash, at $0.500 per million output tokens and a score of 160.4. Context is 1M for Claude Haiku 5.5 and 1.05M for Gemini 3.8 Flash. Use the model pages for every score on file, including lab-reported numbers.
Open this pair in the comparison view
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.