Claude Opus 5 vs Claude Opus 5.5
Claude Opus 5.5 replaces Claude Opus 5 at a lower price and with higher scores on all three benchmarks. Input fell from $5 to $4 per 1M tokens and output from $25 to $20, while Terminal-Bench 4.0 rose from 52.3% to 66.4%. Unless a workflow is pinned to the older model, there is little reason to pick Opus 5 for new work.
What changed from Opus 5 to Opus 5.5
Anthropic released Opus 5 on 24 July 2026 and Opus 5.5 two months later, on 22 September. The limits stayed: 1M tokens of context and 128K tokens of output. The prices did not. Input went from $5 to $4 per 1M tokens, output from $25 to $20, and cached input from $0.50 to $0.20, less than half of what it was.
All three scores went up. Terminal-Bench 4.0 from 52.3% to 66.4% (Opus 5's figure is Anthropic's reproduction; the public leaderboard showed 51.8%, and Opus 5.5 ran at xhigh effort). FrontierCode 1.1 from 48% to 54.4%. Humanity's Last Exam with tools from 63.6% to 67.7%.
What the upgrade saves in a coding session
Our feature session costs $27 on Opus 5 and $21.60 on Opus 5.5. The cached version falls further, from $7.56 to $5.18, because the cache price fell further than the others. A long agent day goes from $244 to $195, or from $61.50 to $41.10 with 90% of prompts cached.
Before you switch
A setting or script that names Opus 5 by its full model id stays on Opus 5 until you change it. Changing the model in Claude Code covers the flag, the setting and the in-session command. Prompts tuned on Opus 5 are worth a run on Opus 5.5 before you move a long job over.
Side by side
| Claude Opus 5 | Claude Opus 5.5 | |
|---|---|---|
| Maker | Anthropic | Anthropic |
| Released | 24 July 2026[1] | 22 September 2026[2] |
| Context window | 1M tokens[1] | 1M tokens[2] |
| Max output | 128K tokens[1] | 128K tokens[2] |
| Input price | $5 per 1M tokens[1] | $4 per 1M tokens[2] |
| Output price | $25 per 1M tokens[1] | $20 per 1M tokens[2] |
| Cached input | $0.50 per 1M tokens[3] | $0.20 per 1M tokens[3] |
| Long prompts | One price | One price |
| Price change | None announced | None announced |
| Open weights | No[1] Offered through the Claude API and cloud platforms; no weights are published. | No[2] Offered through the Claude API and cloud platforms; no weights are published. |
| Reads | text, image[1] | text, image[2] |
Benchmarks
Scores as the makers published them, each with its source and date. Makers run these at different settings, so read a gap of a few points as a tie.
| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| Terminal-Bench 4.0Agentic coding tasks in a real terminal. | 52.3%[4] Anthropic's reproduction; the public leaderboard showed 51.8%. | 66.4%[4] Run at xhigh effort. |
| FrontierCode 1.1 (Main)Hard software engineering tasks. | 48%[4] | 54.4%[4] Max effort. |
| Humanity's Last Exam (with tools)Expert-level questions across many fields, tools allowed. | 63.6%[4] | 67.7%[4] |
What a coding session costs
Our own estimate from the list prices above, for three sessions we define: a coding agent resends the conversation with every call, so prompts are large and grow. "Cached" assumes 90% of prompt tokens are read from the prompt cache and leaves out cache write fees. A long-prompt price applies when a call's prompt is over its threshold. Work out your own mix with the token cost calculator.
| Session | Claude Opus 5 | Claude Opus 5 cached | Claude Opus 5.5 | Claude Opus 5.5 cached |
|---|---|---|---|---|
| Quick fix10 calls of 25K prompt and 1.5K output tokens | $1.63 | $0.61 | $1.30 | $0.45 |
| Feature session60 calls of 80K prompt and 2K output tokens | $27 | $7.56 | $21.60 | $5.18 |
| Long agent day300 calls of 150K prompt and 2.5K output tokens | $244 | $61.50 | $195 | $41.10 |
Value for money and release gap
Claude Opus 5.5 came out 2 months later (60 days). Price per point divides our feature-session cost by a published score: lower means more score for the money. It is a rough lens, not a ranking.
| Measure | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| Input tokens per $1 | 200K | 250K |
| Filling the whole context window once | $5.00 | $4.00 |
| Feature session per point of Terminal-Bench 4.0 | $0.52 | $0.33 |
| Feature session per point of FrontierCode 1.1 (Main) | $0.56 | $0.40 |
| Feature session per point of Humanity's Last Exam (with tools) | $0.42 | $0.32 |
Questions
Is Opus 5.5 cheaper than Opus 5?
Yes. Opus 5.5 costs $4 input and $20 output per 1M tokens; Opus 5 cost $5 and $25. Cached input fell from $0.50 to $0.20.
How much better is Opus 5.5 than Opus 5?
By the published scores, 14.1 points on Terminal-Bench 4.0 (66.4% against 52.3%), 6.4 points on FrontierCode 1.1 and 4.1 points on Humanity's Last Exam with tools.
When did Opus 5.5 come out?
On 22 September 2026, two months after Claude Opus 5, which Anthropic released on 24 July 2026.
Most compared
Any other two models: pick them in the lists at the top of the page.
- Claude Opus 5.5 vs Claude Sonnet 5.5
- Claude Fable 5.1 vs Claude Opus 5.5
- Claude Opus 5.5 vs GPT-6.1 Sol
- Claude Opus 5.5 vs GPT-5.6 Sol
- Claude Opus 5.5 vs GPT-6 Astra
- Claude Opus 5.5 vs Gemini 4 Argon
- Claude Opus 5.5 vs Gemini 3.1 Pro Preview
- Claude Opus 5.5 vs Grok 4.7
- Claude Opus 5.5 vs DeepSeek V4 Pro
- Claude Sonnet 5 vs Claude Sonnet 5.5
- Claude Haiku 5.5 vs Claude Sonnet 5.5
- Claude Haiku 4.5 vs Claude Haiku 5.5
- Claude Fable 5 vs Claude Fable 5.1
- Claude Sonnet 5.5 vs GPT-6.1 Sol
- Claude Fable 5.1 vs GPT-6 Astra
- Claude Haiku 5.5 vs GPT-6 Luna
- GPT-6.1 Sol vs GPT-6 Sol
- GPT-6.1 Sol vs GPT-6 Astra
- Claude Sonnet 5.5 vs Gemini 4 Argon
- Gemini 4 Argon vs GPT-6.1 Sol
- Gemini 3.1 Pro Preview vs Gemini 4 Argon
- Claude Haiku 5.5 vs Gemini 3.8 Flash
- Claude Sonnet 5.5 vs Kimi K3
- Claude Sonnet 5.5 vs GLM-5.3
More tools
Sources
Numbers from the makers' own pages, checked by hand, last read . List prices in US dollars before tax. Spotted a mistake? office@shuriken.sh.
Made by Shuriken, the harness for coding agents. Its hot swapping moves your chats to another of your accounts or API keys when one nears its limit. Use only accounts you own and are entitled to use.
Try Shuriken on your own computerYou're signed in
Add your Claude, Codex and Gemini accounts once. Shuriken switches between them so your agents keep working.Your accounts, projects and chats are in the app.
Coming soon