#1 since Oct 23, 2026. No earlier leader is on record.
Tops major coding benchmarks and leads LMArena for building software and complex debugging. These picks are readable without an account, and the details behind each one are on the card.
LMArena Code 1813; #1 SWE-bench Pro (89.9%) Source.
Cost
Pro is $20/month ($17/month billed annually); API is $4 per million input tokens and $20 per million output tokens.
Free tier
Not on the free plan. Free Claude includes Haiku and Sonnet, not Opus.
Privacy default
Anthropic says Free, Pro, and Max chats are used for training only if you allow it, and you choose at signup. The toggle has been reported as on by default, so check it. Settings, then Privacy. Turn off Help improve Claude. Source
When not to use it
Skip it for simple edits and quick boilerplate where faster, cheaper models suffice.
Starter prompt
Find the bug in this function, explain the cause, and give the fixed version.
LMArena Code 1786.1; 90.2% on SWE-bench Pro V2 HARD Source.
Cost
Plus is $20/month; API is $10 per million input tokens and $50 per million output tokens.
Free tier
Not on the free plan. Plus includes access inside Work and Codex.
Privacy default
Yes on personal Free, Plus, and Pro. Improve the model for everyone is on by default. Settings, Data controls, turn off Improve the model for everyone. Source
When not to use it
Skip it if you want to run completely offline without sending source code to proprietary servers.
Starter prompt
Implement this feature in the repo, then list the files you changed and why.
LMArena Code not cited; 88.2% on SWE-bench Pro V2 HARD. Source.
Cost
Free to download weights; hosted Kimi app is free with usage limits or available via API.
Free tier
Yes, available for free with standard rate limits in the Kimi web and mobile apps.
Privacy default
Yes. The Kimi consumer policy says prompts and files can be used to improve models. Ask Kimi support to stop training use if the app has no toggle. Source
When not to use it
Skip self-hosting if you lack multi-GPU cluster infrastructure required to serve 2.8 trillion parameters.
Starter prompt
Refactor this module for clarity and add tests for the edge cases.
Email
One email when a leader changes. Buttondown is the only service that holds the address, and it uses it only to send these alerts. Open and click tracking stay off. Privacy
Coming soon.
Leader history
Oct 23, 2026. Claude Opus 5.5 is the first leader on record. LMArena Code 1813; #1 SWE-bench Pro (89.9%)
A new number one has to show up on two daily checks before it replaces the current leader. Get alerts.
Source: WhichAI.today, Coding leader as of Oct 24, 2026, whichai.today/coding
Questions
Which AI is best for coding & software right now?
Claude Opus 5.5 from Anthropic leads, as of Oct 24, 2026. Tops major coding benchmarks and leads LMArena for building software and complex debugging. The alternative is GPT-6 Astra, and the open-weights pick is Kimi K3.
Where do I get Claude Opus 5.5?
Claude app, Pro plan ($20/month), choose Opus 5.5. Pro is $20/month ($17/month billed annually); API is $4 per million input tokens and $20 per million output tokens.
Has this leader changed?
Tracking for this category began on Oct 23, 2026, with Claude Opus 5.5 as the first recorded leader.