Claude Opus 5: release date and what it's for
Released , according to Anthropic's announcement. Legacy since Opus 5.5 shipped on 22 September. Opus 5.5 costs less per token, so pinning here only makes sense if a prompt regressed.
- Status
- Live
- Released
- Lab
- Anthropic
- Input
- $5per million tokens
- Output
- $25per million tokens
- Context window
- 1Mtokens
- Max output
- 128Ktokens
API model ID: claude-opus-5
What is Claude Opus 5 capable of?
6 posts from people showing what they made with Claude Opus 5, newest and most-reacted first. People are posting video, music, data and game made with it.
Each card is the original post, shown the way X or Reddit embeds it, with a link to where it came from. We picked the ones that show something being made and skipped the benchmark screenshots. We can't vouch for any of them being unedited.
What we've written about Claude Opus 5
Goodfire says its cyber monitors for open models beat an AI judge at 50x lower cost
Cheap enough to run on every token, the probes can stop an agent before it touches a tool.
Mistral previews Large 4, a 1 trillion-parameter model, and says weights come by month's end
Mistral says Claude Opus 5.5 and GPT-6 Astra refuse a cyber test that ML4 scores 82% on.
Claude models explained: Opus 5.5, Fable 5.1, Mythos and why there's no Opus 5.1
The newest Opus beats the flagship on Anthropic's own benchmark table, but the flagship is still the one Anthropic tells you to move up to.
Opus 5.5 sends most cybersecurity tasks to Opus 4.8, and costs 20% less per token
Anthropic's first model since the pacing essay ships with Fable 5.1's biology and cyber safeguards, and it says the model often suspects it's being tested.
Xiaomi's MiMo-V2.6-Pro tops the open-weights index, but its own table puts Opus 5 ahead
Xiaomi trained it with one reinforcement learning run across coding, agents, vision and cybersecurity, and put the weights on Hugging Face under MIT.
Opus 4.8 couldn't write a working exploit, but Opus 5 did it in three hours
The chain it wrote reached OpenAI employees' ChatGPT accounts, and from there a pull request in the private openai/openai repo.
The day after agreeing AI should slow down, Musk said Grok 4.8 finishes training this week
Grok 4.7 still isn't out, Grok 5 is his answer for AGI, and neither OpenAI nor Anthropic lists a size to set 2.5T against.
We priced a home LLM rig against a $20 subscription. It breaks even in 23 years.
Every number below is from the labs' own pricing pages and docs, checked on September 5, 2026.
How Claude Opus 5 compares on price and context
inputoutput
- Claude Mythos 5.1$10 / $50
- Claude Fable 5.1$10 / $50
- GPT-6 Astra$10 / $50
- Claude Opus 5$5 / $25
- Claude Opus 5.5$4 / $20
- Muse Spark 1.3$1.25 / $4.25
- Gemini 3.8 Flash$0.75 / $3.75
how much it reads at once
- GPT-6 Astra1.05M
- Muse Spark 1.31.05M
- Gemini 3.8 Flash1.05M
- Claude Opus 51M
- Claude Opus 5.51M
- Claude Mythos 5.11M
- Claude Fable 5.11M
Next to the newest Anthropic models and the current model from other labs. Prices are in US dollars per million tokens.
What is Claude Opus 5?
Claude Opus 5 is a model from Anthropic, built in the US. It's out, and you can use it today.
When did Claude Opus 5 come out?
Anthropic released it on July 24, 2026. That date comes from Anthropic's announcement.
More from Anthropic
Claude Opus 5.5
- in /1M
- $4
- out /1M
- $20
- context
- 1M
Claude Mythos 5.1
- in /1M
- $10
- out /1M
- $50
- context
- 1M
Claude Fable 5.1
- in /1M
- $10
- out /1M
- $50
- context
- 1M
Claude Sonnet 5
- in /1M
- $2
- out /1M
- $10
- context
- 1M
Claude Fable 5
- in /1M
- $10
- out /1M
- $50
- context
- 1M
Claude Haiku 4.5
- in /1M
- $1
- out /1M
- $5
- context
- 200K
Claude Sonnet 4.5
- in /1M
- $3
- out /1M
- $15
- context
- 200K
More news: Anthropic news and Claude updates
.