Claude Sonnet 5.5 vs GPT-6 Sol pricing
Side-by-side LLM API pricing.Input rates are equal ($2.00 vs $2.00 /1M), and output rates are equal ($10.00 vs $10.00 /1M).
| Attribute | Claude Sonnet 5.5 | GPT-6 Sol |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Tier | Mid-range | Flagship |
| Status | GA | GA |
| Context window | 1M | 1.05M |
| Input /1M | $2.00 | $2.00 |
| Output /1M | $10.00 | $10.00 |
| Cached input /1M | $0.20 | $0.20 |
| Batch input /1M | $1.00 | $1.00 |
| Batch output /1M | $5.00 | $5.00 |
| Blended (avg in+out) | $6.00 | $6.00 |
| 1M in + 1M out | $12.00 | $12.00 |
| 1M in (90% cached) + 100K out | $1.38 | $1.38 |
List prices, USD, directional. Rates are provider list prices per 1M tokens and are meant for comparison, not billing. Batch rates shown as 50% off are derived where a provider offers batch but does not publish a separate figure. Preview, promo, intro, peak/off-peak, long-context, and third-party-host prices are labeled where they apply. Token counts vary by tokenizer, so per-token price is not always a like-for-like cost. Always confirm with the provider before relying on a number. Prices as of 2026-09-29.
Tokenizer caveat: Claude Sonnet 5.5 uses a tokenizer that produces more tokens per unit of text, so identical content is billed as a different number of tokens on each model. Per-token price is therefore not a like-for-like cost — compare cost per task, not just the sticker rate.
Which should you pick?
For a balanced job (1M input + 1M output), both models cost $12.00. If your prompts are large and reused, prompt caching changes the maths — the cache-heavy row above shows a 90%-cached input scenario. If your work can run asynchronously, both providers' batch rates cut the bill further where offered.
Full price matrices: Claude Sonnet 5.5 pricing → · GPT-6 Sol pricing → · Back to the LLM pricing hub →