Gemini vs Claude API Pricing: gemini-2.5-pro vs Sonnet 4.6
gemini-2.5-pro is cheaper on paper ($1.25/$10 per 1M vs Sonnet 4.6's $3/$15) and adds full multimodality, but Gemini prices rise above a 200K-token prompt and output includes thinking tokens. Choose Gemini for multimodal and large-doc value; choose Sonnet for coding and agent reliability.
Choose gemini-2.5-pro for multimodal input, large-document analysis, and lower headline token prices.
Choose Claude Sonnet 4.6 for coding, agents, and predictable flat token pricing across a 1M context.
Side by side
| Gemini (2.5 Pro) | Claude (Sonnet 4.6) | |
|---|---|---|
| Input | $1.25/1M (≤200K) | $3.00/1M |
| Output | $10.00/1M (≤200K) | $15.00/1M |
| Above 200K prompt | $2.50 in / $15 out | Flat $3 / $15 |
| Cached input | $0.125/1M + storage | $0.30/1M read |
| Modality | Text/image/video/audio/PDF | Text + vision/PDF |
Gemini 2.5 Pro undercuts Claude Sonnet 4.6 on headline price — $1.25/$10 versus $3.00/$15 per 1M — and adds full multimodal input across image, video, audio, and PDF. For large-document and multimodal workloads, that's strong value.
Two caveats matter. Gemini's pricing tiers up once a prompt exceeds 200K tokens (to $2.50 input / $15 output), and output billing includes thinking tokens, so reasoning-heavy responses cost more than they appear. Claude Sonnet's flat $3/$15 across a 1M window is more predictable, and it's a favorite for coding and agents.
FAQ
Is Gemini cheaper than Claude?
Which is better for coding?
Related comparisons
GPT (gpt-5.5) vs Claude (Opus/Sonnet)
For flagships, gpt-5.5 ($5/$30 per 1M) and Claude Opus 4.8 ($5/$25) are close, with Opus cheaper on output. Fo…
OpenAI API vs Anthropic API
Both are frontier API providers with cached-input and batch discounts. OpenAI's gpt-5.5 outputs at $30/1M; Ant…
Claude API vs AWS Bedrock Claude
Use the direct Anthropic API for transparent token pricing with explicit cache-read/write and batch tiers (Son…