Claude Opus 5 Complete Guide
Same price and the same context window as Opus 4.8, so you can switch freely. The real difference is prompt caching: the threshold drops from 2048 tokens to 512, which brings short system prompts and tool definitions into cache range.
Both claude-opus-5 and claude-opus-5[1m] are open — no application, no surcharge. Opus 4.8 remains in service, and since the two share identical pricing and context limits you can move between them as you like.
The numbers
$5 in, $25 out — the same as Opus 4.8. QCode's Claude service rate is x1.0
1M tokens, enabled explicitly with claude-opus-5[1m]
Cacheable from 512 tokens; Opus 4.8 needs 2048
Up to 128K tokens in a single response
What actually differs from Opus 4.8
The short answer: price, context window and capability set are identical. The one substantive difference is the minimum length for prompt caching.
Price and window: identical
$5 in / $25 out per million tokens, $6.25 cache write and $0.50 cache read, 1M input and 128K output — line for line the same on both models. Switching therefore produces no billing surprises and needs no fresh cost modelling.
Prompt cache threshold: 512 vs 2048
This is the only substantive difference. Prompt caching has a minimum length, below which content is simply not cached. Opus 5 lowers that floor to 512 tokens — a quarter of Opus 4.8's.
Cache pricing is the same on both ($6.25 write, $0.50 read per million tokens). What changes is only where the line sits for "long enough to be worth caching". Agentic workflows with a short system prompt, many tool definitions and dense turns are where this is most noticeable.
Capabilities: the same set
Vision input, function calling, PDF input, computer use and prompt caching are flagged identically on both models. Integration code written for Opus 4.8 needs no changes.
How to call it
Switch inside Claude Code
Use the /model command mid-session; no restart required.
/model claude-opus-5Set it as the default via environment variable
Useful when you always want one model — put it in your shell profile to make it stick.
ANTHROPIC_MODEL=claude-opus-5 claudeEnable the 1M token context
The [1m] suffix is how you explicitly ask for the long context window.
/model claude-opus-5[1m]Choosing across the Claude lineup
Official list prices per million tokens. QCode's Claude service rate is x1.0.
| Model | Input $/M | Output $/M | Context / output | Best for |
|---|---|---|---|---|
| claude-opus-5 | $5.00 | $25.00 | 1M / 128K | The new flagship; the lower cache floor suits tool-heavy, many-turn agentic work |
| claude-opus-4-8 | $5.00 | $25.00 | 1M / 128K | Same price and window, and still the platform's main workhorse today |
| claude-fable-5 | $10.00 | $50.00 | 1M / 128K | Priced above Opus, aimed at long-form and creative writing |
| claude-sonnet-5 | $2.00 | $10.00 | 1M / 128K | The value pick for everyday coding |
| claude-haiku-4-5 | $1.00 | $5.00 | 200K / 64K | Formatting, simple Q&A, bulk operations |
Pricing comes from the LiteLLM model pricing database, the same live source as our models page. Full list on the models and pricing page
Frequently asked questions
Should I use Opus 5 or Opus 4.8?
They cost the same and share a context window, so neither choice is the expensive mistake. If your workflow has many tool definitions and relatively short turns, Opus 5's lower cache threshold may give you better cache hit rates; otherwise pick whichever you are used to. Opus 4.8 has not been retired and remains in service.
Do I need code changes to migrate from Opus 4.8?
No. The capability flags (vision, function calling, PDF, computer use, prompt caching) are identical, so it is a matter of changing claude-opus-4-8 to claude-opus-5. Pricing is unchanged, so your cost estimates still hold.
Which model names are available?
Two: claude-opus-5 and claude-opus-5[1m], the latter to explicitly enable the 1M-token input window. Please use these names; date-suffixed variants are not part of the published set.
What does it cost, including caching?
$5 input and $25 output per million tokens; prompt cache writes are $6.25 and reads $0.50 per million. QCode's Claude service rate is x1.0, i.e. official list price. The models page carries the live figures.
What capabilities does it support?
Vision input, function calling, PDF input, computer use and prompt caching — the same set as Opus 4.8.
Does it need to be enabled separately?
No. Every plan can call every model listed here. Model access is identical across plans, with no unlock fee and no application.
Using Opus 5 on QCode
One key, three platforms
The same API key works in Claude Code, the OpenAI Codex CLI and Gemini clients — no separate purchases.
Every model open
All plans can call all models. Access is not tiered by plan and there is no unlock fee.
Pricing from the official source
Per-token prices come from the LiteLLM pricing database and stay in sync; the Claude service rate is x1.0.
Direct access from mainland China
Usable directly on mainland networks, with no network workarounds needed on your side.
Try Claude Opus 5 now
Sign up and use one key across the full Claude, Codex and Gemini lineup.