Which model is best for coding?

Claude Opus 5.5 vs GPT-6.1 Sol vs GPT-6 Astra vs Claude Sonnet 5.5 vs Mistral Large 4, compared on the jobs that matter for this question. The verdict comes straight from the scores, so it changes when they do. 100 is the best available today in that job.

New to this? What the scores and labels mean

0 to 100: Our score for one job, like writing or coding, compared with everything else out today. 100 is the best available today. For AI models we build it from public data, such as Arena's blind votes and Epoch AI's independent tests; apps and tools are scored on our own checklist. The same tool can score differently for different jobs, and where two likely ranges overlap, we call it a tie.
Our verdict: Use it: a top choice for most people. Worth trying: good, and worth a go. Depends: good for some people or one kind of job. Wait: too early to tell, or not out for everyone yet. Skip: something else does the job better.
First look: Just released. We say what it is and whether to bother, and give it a score once public results exist.
No score yet: None of the public sources we use covers it yet, or the only evidence is from a source whose numbers we can't use, so we describe it in words instead.
Best buy: Nothing cheaper scores as well for that job.
Best free: The best score for that job that costs nothing, through a free plan or by running on your own computer.
Pay as you go: No monthly plan: you pay for each use, usually through a developer account.
Checked: The day we last confirmed the score and the price against the sources. Prices change often, so check the date.
AI model: The engine underneath, like GPT or Claude. You use it through an app or, if you code, directly.
Open source: Free to download and run on your own computer, so nothing leaves your machine.

How we score, in full

Our verdict

Overall, Claude Opus 5.5: the best or tied best in both jobs asked about.

For coding, Claude Opus 5.5, Claude Sonnet 5.5 and GPT-6 Astra are tied (100, 89 and 78); about the same price ($20 a month, free and $20 a month).
For agents, Claude Opus 5.5 (#1 on Arena's Agent board, $20 a month).
On a budget, Claude Sonnet 5.5: 89 in coding for free.

Coding tools such as Claude Code, Codex and Cursor run on these models. They are on the list, and get reviewed on their own soon.

Best marks the top score in each job; when likely ranges overlap, each tied one says Tied. Open in Compare to add or swap items.

Claude Opus 5.5 vs GPT-6.1 Sol vs GPT-6 Astra vs Claude Sonnet 5.5 vs Mistral Large 4: scores by job, price and where it runs
	Claude Opus 5.5Anthropic · AI modelUse it	GPT-6.1 SolOpenAI · AI modelUse it	GPT-6 AstraOpenAI · AI modelDepends	Claude Sonnet 5.5Anthropic · AI modelUse it	Mistral Large 4Mistral AI · AI modelWait
CodingWriting and fixing code	100Tiedlikely 89–100	85likely 81–88	78Tiedlikely 65–92	89Tiedlikely 85–92	36likely 34–38
AgentsAI that does multi-step tasks for you	#1BestArena's Agent board	#4Arena's Agent board	#2Arena's Agent board	#5Arena's Agent board	#39Arena's Agent board
WritingEmails, essays, editing	100Bestlikely 96–100	83likely 78–87	80likely 77–84	85likely 80–89	59likely 55–63
ResearchQuestions, sources, long PDFs	94Tiedlikely 88–99	94Tiedlikely 87–99	93Tiedlikely 88–96	90Tiedlikely 83–94	62likely 56–67
Price for these scoresThe plan the scores above are for	$20 a monthChecked 11 Oct	$20 a monthChecked 11 Oct	$20 a monthChecked 11 Oct	Free or $20 a monthdepends on the jobChecked 11 Oct	Pay as you goChecked 11 Oct
Free to try	No	No	No	Yes	No
Open source	No	No	No	No	Yes
In a line	Top public pick for coding, agents and hard questions at a mid price; skip it for bulk work where Sonnet 5.5 is enough.	Near-top coding and agent work at a fifth of GPT-6 Astra's price; pick Claude Opus 5.5 or Gemini for creative writing.	Second on WebDev and Agent boards, but GPT-6.1 Sol is close for a fifth of the price; middling at creative writing.	Near-top coding, agents and research for half Opus 5.5's token price; for writing, pick Opus 5.5 or Gemini 3.8 Flash.	Mistral's 1T preview scores no higher than 72 in any job; weights due 31 Oct. Use GLM-5.3 or Kimi K3 now.
All five	Cloud only

Other head-to-heads

Which chat app should I use every day?ChatGPT vs Claude vs Gemini
How do I run AI on my own computer?Ollama vs LM Studio (with Bionic agent)
Which AI makes the best images?Nano Banana 2.1 vs GPT Image 2.5 Sunburst vs GPT Image 2.5 Flare vs FLUX 3 Image vs Qwen-Image-2.1 vs Midjourney V8.2
Which AI voice sounds best?Eleven v4 vs Eleven v4 Turbo vs Gemini 3.8 Flash TTS vs Gemini 3.8 Flash-Lite TTS vs GPT-Live 1