Anthropic's and OpenAI's mid-tier models launched a day apart in September 2026 at exactly the same input, output, cache-hit and cache-write prices. On short prompts the bill is identical; the differences appear above 272K tokens and in how caches expire.
On prompts under 272K tokens, Claude Sonnet 5.5 and GPT-6.1 Sol cost the same on every line. Above 272K, GPT-6.1 Sol moves to a higher rate while Sonnet 5.5 keeps standard pricing across its full window. Otherwise, choose on quality, tooling and cache-lifetime fit.
| Claude Sonnet 5.5 | GPT-6.1 Sol | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Tier | Mid-tier | Mid-tier |
| Input / 1M tokens | $2.00 | $2.00 |
| Output / 1M tokens | $10.00 | $10.00 |
| Cache hit / 1M | $0.10 | $0.10 |
| Cache write / 1M | $2.50 | $2.50 |
| Context window | 1M | 1M |
| Notes | Released Sep 28, 2026, replacing Sonnet 5 at the identical $2/$10 per 1M price; Sonnet 5 is now a legacy model | New (Sep 29, 2026 DevDay), below Astra in the GPT-6 family; replaces GPT-6 Sol at the same input/output price with a 50% cheaper cache-read rate; >272K tokens: $4/$15, cache $0.20. ChatGPT Work/Codex on Plus/Pro/Business/Enterprise/Edu, plus API |
Both charge $2.00 per 1M input tokens and $10.00 per 1M output tokens.
| Workload | Claude Sonnet 5.5 | GPT-6.1 Sol | Difference |
|---|---|---|---|
| Support chatbot 10,000 conversations × (2,000 input + 400 output tokens) | $80.00 | $80.00 | — |
| Document Q&A 1,000 questions × (30,000 input + 800 output tokens) | $68.00 | $68.00 | — |
| Coding agent session 4M input tokens (90% cache hits) + 150,000 output tokens | $2.66 | $2.66 | — |
Standard rates per 1M tokens from data on this site. The coding-agent row assumes 90% of input is read from the prompt cache (full input price where a model has no cache rate). Cache-write fees, batch discounts and long-context surcharges listed in the notes are not included. Updated November 2, 2026.
you send prompts above 272K tokens — whole repos, long contracts — where Sonnet 5.5 keeps its standard rate, or you prefer Anthropic's 1-hour cache option.
you are on OpenAI's platform, want caches that live 30 minutes by default (or 12 hours), or want the half-price Flex tier for latency-tolerant traffic.
GPT-6.1 Sol's long-context band (requests over 272K tokens) is 2× input, 1.5× output and 2× cache hits. Anthropic's pricing page states Sonnet 5.5 includes the full 1M window at standard pricing. If even a minority of your requests are that long, Sonnet's bill comes out lower.
Both charge $2.50 to write and $0.10 to read 1M cached tokens. But Anthropic's write buys a 5-minute cache (a 1-hour write costs 2× input), while OpenAI's buys 30 minutes, with a 12-hour option at a higher write rate. Assistants with long gaps between turns may re-write the cache more often on Anthropic's default TTL — use the 1-hour write if users pause.
Both halve input and output prices for Batch jobs. OpenAI also offers Flex at half price for slower, interruptible traffic, and a Fast tier at a premium. Region pinning costs extra on both: 1.1× for Anthropic's US-only inference, a 10% uplift for OpenAI's regional processing on eligible models.
Yes on standard rates: $2.00 input, $10.00 output, $0.10 cache hit and $2.50 cache write per 1M tokens on both. GPT-6.1 Sol costs more above 272K tokens.
No. Anthropic bills its full 1M window at standard rates.
Anthropic's standard cache write lasts 5 minutes (1-hour option available); OpenAI's lasts 30 minutes, with a 12-hour option.
Both Batch APIs are 50% off. OpenAI also has a half-price Flex tier.
How we choose: prices come from each vendor's official pricing page and are re-checked every week. We take no affiliate fees, so rankings aren't for sale.