Model comparison
Claude Opus 5.5 vs GPT-5.6 Sol: same list price, $0.20 apart
Claude Opus 5.5 and GPT-5.6 Sol share $4 input and $20 output prices. Cache rules decide the rest: a coding session costs $4.40 on Opus 5.5 and $4.20 on Sol.
· Prices as of September 28, 2026
Claude Opus 5.5
Anthropic · Released September 22, 2026
Anthropic's recommended starting model for most work, built for long-running agentic coding. Anthropic says it matches Claude Fable 5.1 on most work at a much lower price.
Claude Opus 5.5 facts and comparisonsGPT-5.6 Sol
OpenAI · Released July 9, 2026 · Previous generation
The flagship of the July 2026 GPT-5.6 family, now succeeded by GPT-6 Sol and on promotional pricing.
GPT-5.6 Sol facts and comparisons
The short answer
Claude Opus 5.5 and GPT-5.6 Sol have the same list prices, $4 input and $20 output per million tokens, so uncached work costs the same on both. On the example agentic coding session GPT-5.6 Sol is $0.20 cheaper, $4.20 against $4.40, because Anthropic's 1-hour cache writes cost more than Opus 5.5's cheaper cache hits save. Choose Opus 5.5 in Claude Code, especially on the 5-minute cache, and GPT-5.6 Sol only where you already use it, since it runs on promotional rates and Codex now suggests GPT-6 Sol.
Choose Claude Opus 5.5 if
- You use Claude Code with an API key, where the 5-minute cache writes at $5, the same as GPT-5.6 Sol, and each hit costs half as much.
- You want a current model: Opus 5.5 was released on September 22, 2026 and is Claude Code's default.
- Your jobs are long and sprawling, like codebase-wide migrations and audits, which Anthropic names as a strength.
- You send prompts over 272K input tokens, where GPT-5.6 Sol's price rises and Opus 5.5 stays at standard rates up to 1M.
Choose GPT-5.6 Sol if
- You chat in Codex cloud on a ChatGPT plan, which runs on GPT-5.6 Sol.
- Your work is frontend-heavy, and OpenAI claims better layout, visual hierarchy, and design judgment for the GPT-5.6 family.
- Your code already calls the gpt-5.6 API id, which points to this model.
Side by side
Specs and prices
| Fact | Claude Opus 5.5 | GPT-5.6 Sol |
|---|---|---|
| Maker | Anthropic | OpenAI |
| API model id | claude-opus-5-5 | gpt-5.6-sol |
| Released | September 22, 2026 | July 9, 2026 |
| Status | Current | Previous generation |
| Context window | 1M tokens | 1.05M tokens |
| Max output | 128K tokens | 128K tokens |
| Open weights | No | No |
| Input, per 1M tokens | $4 | $4 |
| Cache hit, per 1M | $0.20 | $0.40 |
| Cache write, per 1M | $5 (5-minute), $8 (1-hour) | $5 |
| Output, per 1M tokens | $20 | $20 |
| Runs in | Claude Code, Cursor, OpenCode, OpenRouter, and GitHub Copilot | Codex, Cursor, OpenCode, OpenRouter, and GitHub Copilot |
Standard API rates in US dollars, as published by each maker (Claude Opus 5.5: September 26, 2026; GPT-5.6 Sol: September 28, 2026). Batch and priority tiers, taxes, and subscription plans are not included. Claude Opus 5.5: The full 1M context window is billed at standard rates. Fast mode, a research preview on the Claude API, costs $8 input and $40 output per million tokens. GPT-5.6 Sol: Requests over 272K input tokens cost 2x for input and cache and 1.5x for output, for the whole request. These are promotional rates, available at least through November 21, 2026.
Cost
What the same work costs
The same token counts, priced at each model’s published rates. Cached tokens are billed at each maker’s cache prices, so the session shows what caching is worth on each model.
Real sessions differ: the two models count the same code as different numbers of tokens, and reasoning settings change how much each one writes. Your own history is the real test.
| Workload | Claude Opus 5.5 | GPT-5.6 Sol |
|---|---|---|
| Agentic coding session, 100K input, 400K written to cache, 2M read from cache, 50K output | $4.40 | $4.20 |
| Large one-off review, 150K input with no cache hits, 10K output | $0.80 | $0.80 |
| Output-heavy generation, 30K input, 80K output | $1.72 | $1.72 |
| A month of sessions, 110 sessions: 5 a day, 22 working days | $484.00 | $462.00 |
| Where the session’s cost goes | ||
| Cache writes | $2.60 | $2.00 |
| Cache reads | $0.40 | $0.80 |
| Uncached input | $0.40 | $0.40 |
| Output | $1.00 | $1.00 |
- caching saves on the session with Claude Opus 5.5 (60%)
- $6.60
- caching saves on the session with GPT-5.6 Sol (62%)
- $6.80
Identical list prices, different cache rules
Claude Opus 5.5 and GPT-5.6 Sol charge $4 per million input tokens, $20 per million output tokens, and $5 for a standard cache write, which both makers set at 1.25x input. With no caching involved they cost exactly the same: $0.80 for the large one-off review and $1.72 for output-heavy generation.
Two caching rules differ. Anthropic prices an Opus 5.5 cache hit at 0.05x input, $0.20 per million, half of GPT-5.6 Sol's $0.40 at OpenAI's 0.1x. Anthropic also sells a 1-hour write at 2x input, $8 per million, which OpenAI does not; on GPT-5.6 and later, a cached prefix instead stays reusable for at least 30 minutes after its last use.
The $0.20 that separates them on a session
In the example session, Opus 5.5's cheaper hits save $0.40: 2M cached tokens cost $0.40 on it and $0.80 on GPT-5.6 Sol. Its 1-hour writes cost $0.60 more, $2.60 in writes against $2.00. Net, GPT-5.6 Sol comes out $0.20 ahead, $4.20 against $4.40, or $22.00 over 110 sessions a month.
That margin depends on which Anthropic cache you use. Claude Code puts the main conversation on the 1-hour cache with a Claude subscription and on the 5-minute cache with an API key. On the 5-minute cache, Opus 5.5 writes at the same $5 as GPT-5.6 Sol and keeps its cheaper reads, so it comes out ahead. Anthropic's rule of thumb is that a 1-hour write pays for itself after two reads.
Either way, the gap is small next to the effect of reasoning settings and tokenizers. Opus 5.5 uses Anthropic's newer tokenizer, which counts about 30% more tokens than earlier Claude models for the same text, and OpenAI says GPT-5.6 reaches flagship-level performance with fewer output tokens.
A current model against one on promotion
GPT-5.6 Sol launched on July 9, 2026 as the flagship of the GPT-5.6 family. It is now a previous-generation model on promotional rates, available at least through November 21, 2026. OpenAI has released GPT-6 Sol as its successor, and Codex suggests moving to it everywhere except Codex cloud chats on ChatGPT plans, which still use GPT-5.6 Sol.
Opus 5.5 is Anthropic's current recommended starting model. Anthropic says it "performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5," and offers fast mode on the Claude API at $8 input and $40 output as a research preview. Both models write up to 128K tokens, with context windows of 1M on Opus 5.5 and 1.05M on GPT-5.6 Sol.
To check which one costs less on your own work, EveryToken reads your Claude Code, Codex, and Cursor history on a Mac and prices each request at API rates, with cache savings shown per model.
Prompt caching
How each maker bills cached tokens
Anthropic
Claude caches a prompt prefix up to a breakpoint. One top-level cache_control field places the breakpoint automatically and moves it as the conversation grows, or you can mark up to 4 blocks yourself. Claude Code manages caching for you.
A 5-minute cache write costs 1.25x the input price and a 1-hour write costs 2x. A cache hit costs 0.1x input on most models, 0.05x on Claude Opus 5.5, and 0.025x on Claude Fable 5.1 and Claude Mythos 5.1. Every hit restarts the cache lifetime at no charge.
Anthropic's rule of thumb: a 5-minute write pays for itself after one cache read, and a 1-hour write after two.
In Claude Code, the main conversation uses the 1-hour cache on a Claude subscription and the 5-minute cache with an API key. Each model has its own cache, so switching models starts over.
Source: Anthropic: Prompt caching
OpenAI
Prompt caching is on by default. From GPT-5.6 on you can also mark up to four explicit cache breakpoints, while GPT-5.5 and earlier cache automatically only.
From GPT-5.6 on, a cache write costs 1.25x the uncached input price and a cache hit costs 0.1x. GPT-5.5 and earlier add no charge for writing the cache: written tokens are billed as ordinary input, and a hit costs 0.1x on the models compared here.
On GPT-5.6 and later, a cached prefix stays reusable for at least 30 minutes after its last use, and caching starts at 1,024 input tokens.
Source: OpenAI: Prompt caching
Your own numbers
See what Claude Opus 5.5 and GPT-5.6 Sol really cost you.
everyaitoken reads your Claude Code, Cursor, OpenCode, OpenRouter, and Codex history on your Mac and prices every request at API rates, with what caching saved or cost. $9 once.
FAQ
Questions
Do Claude Opus 5.5 and GPT-5.6 Sol cost the same?
Their list prices match: $4 input, $20 output, and $5 per million for a standard cache write. Uncached work costs the same on both. Cached sessions differ slightly, because Opus 5.5 has cheaper hits and a pricier 1-hour write.
Why is GPT-5.6 Sol cheaper on the cached session?
Half of the session's cache writes use Anthropic's 1-hour cache at $8 per million, which costs Opus 5.5 $0.60 more than GPT-5.6 Sol's writes. Opus 5.5's cheaper hits win back $0.40 of that, leaving GPT-5.6 Sol $0.20 cheaper.
What happens to GPT-5.6 Sol's price after November 21, 2026?
OpenAI lists its current rates as promotional and available at least through that date. The sources behind this page do not give the price after it, so recheck before relying on these figures later in the year.
Where can I use each model?
Both run in Cursor, OpenCode, OpenRouter, and GitHub Copilot. Opus 5.5 is Claude Code's default, and GPT-5.6 Sol runs in Codex.
Sources
- Anthropic: Pricing
- Anthropic docs: Claude Opus 5.5
- Anthropic: Claude Opus 5.5
- Anthropic docs: Fast mode
- Claude Code docs: Model configuration
- Cursor docs: Claude Opus 5.5
- OpenRouter: Claude Opus 5.5
- OpenCode docs: Zen
- GitHub Docs: Supported AI models in Copilot
- OpenAI: API pricing
- OpenAI docs: GPT-5.6 Sol
- OpenAI: Using GPT-5.6
- OpenAI: API changelog
- Codex docs: Models
- Codex docs: Pricing
- Cursor docs: Models and pricing
- OpenRouter: GPT-5.6 Sol
- Anthropic: Prompt caching
- OpenAI: Prompt caching