Live ranking, refreshed every 2 days
Best AI Models for Coding and Web Development
As of October 8, 2026, the top-ranked model on LMArena's WebDev arena is claude-opus-5.5-max by Anthropic with a score of 1,813, 27 points ahead of gpt-6-astra-max by OpenAI.
Updated October 13, 2026. Showing the last saved copy of the leaderboard, dated October 8, 2026, because the live source could not be reached.
Ranking: top 10 models
Scores come from blind head-to-head votes on LMArena's WebDev arena. Rank is LMArena's published rank, which accounts for confidence intervals, so several models can share a rank. The top 10 are shown below, and ranks 11 to 40 follow in the expandable list.
| Rank | Model | Lab | Score | Votes | License |
|---|---|---|---|---|---|
| 1 | claude-opus-5.5-max | Anthropic | 1,813 ±15 | 2,294 | Proprietary |
| 2 | gpt-6-astra-max | OpenAI | 1,786 ±11 | 5,189 | Proprietary |
| 3 | claude-sonnet-5.5-xhigh | Anthropic | 1,774 ±15 | 2,323 | Proprietary |
| 4 | gpt-6.1-sol-max | OpenAI | 1,755 ±15 | 2,102 | Proprietary |
| 5 | claude-fable-5.1-max | Anthropic | 1,744 ±9 | 6,663 | Proprietary |
| 6 | claude-sonnet-5.5-high | Anthropic | 1,715 ±11 | 3,984 | Proprietary |
| 7 | claude-opus-5-max | Anthropic | 1,691 ±6 | 17,320 | Proprietary |
| 8 | gpt-6-sol-max | OpenAI | 1,688 ±12 | 3,407 | Proprietary |
| 9 | gemini-4-argon-high | 1,678 ±15 | 1,771 | Proprietary | |
| 10 | qwen3.8-max-0902 | Alibaba (Qwen) | 1,674 ±8 | 11,182 | Proprietary |
Show ranks 11 to 40
Rank Model Lab Score License 11 qwen3.8-max Alibaba (Qwen) 1,672 ±12 Proprietary 12 claude-opus-5-high Anthropic 1,657 ±6 Proprietary 13 muse-spark-1.3-max Meta 1,657 ±8 Proprietary 14 kimi-k3-max Moonshot AI 1,654 ±6 Open weights 15 grok-4.7-xhigh xAI 1,639 ±11 Proprietary 16 mimo-v2.6-flash Xiaomi 1,637 ±10 Open weights 17 qwen3.8-flash-next Alibaba (Qwen) 1,633 ±9 Open weights 18 hy4-preview Tencent 1,632 ±10 Open weights 19 mimo-v2.6-pro Xiaomi 1,629 ±11 Open weights 20 muse-spark-1.3 (xHigh) Meta 1,626 ±8 Proprietary 21 claude-fable-5-high Anthropic 1,625 ±6 Proprietary 22 glm-5.3-max Z.ai (Zhipu) 1,622 ±8 Open weights 23 deepseek-v4.1-flash-max DeepSeek 1,619 ±11 Open weights 24 gpt-5.6-sol-xhigh (codex-harness) OpenAI 1,618 ±6 Proprietary 25 grok-4.6-high xAI 1,617 ±8 Proprietary 26 glm-5.3-flash Z.ai (Zhipu) 1,609 ±7 Open weights 27 glm-5.2-max Z.ai (Zhipu) 1,603 ±6 Open weights 28 qwen3.8-27b Alibaba (Qwen) 1,593 ±7 Open weights 29 gemini-3.7-flash-high Google 1,592 ±7 Proprietary 30 claude-haiku-5.5-high Anthropic 1,587 ±20 Proprietary 31 gemini-3.8-flash-high Google 1,583 ±8 Proprietary 32 deepseek-v4-pro-high-20260813 DeepSeek 1,582 ±9 Open weights 33 deepseek-v4-flash-high DeepSeek 1,582 ±9 Open weights 34 gpt-6-luna-max OpenAI 1,581 ±7 Proprietary 35 step-5-preview-high StepFun 1,564 ±11 Proprietary 36 claude-opus-4-7-high Anthropic 1,557 ±6 Proprietary 37 claude-opus-4-8-high Anthropic 1,556 ±6 Proprietary 38 claude-opus-4-7 Anthropic 1,555 ±6 Proprietary 39 grok-4.5 xAI 1,553 ±7 Proprietary 40 claude-opus-4-6-high Anthropic 1,547 ±6 Proprietary
Showing the top 40 of 141 models on this leaderboard. Data from October 8, 2026.
Best model by lab
Each lab's highest-ranked model on this leaderboard, and how far it sits from the top.
| Lab | Best model | Rank | Score | Behind #1 |
|---|---|---|---|---|
| Anthropic | claude-opus-5.5-max | 1 | 1,813 | Leader |
| OpenAI | gpt-6-astra-max | 2 | 1,786 | 27 pts |
| gemini-4-argon-high | 9 | 1,678 | 135 pts | |
| Alibaba (Qwen) | qwen3.8-max-0902 | 10 | 1,674 | 140 pts |
| Meta | muse-spark-1.3-max | 13 | 1,657 | 156 pts |
| Moonshot AI | kimi-k3-max | 14 | 1,654 | 159 pts |
| xAI | grok-4.7-xhigh | 15 | 1,639 | 174 pts |
| Xiaomi | mimo-v2.6-flash | 16 | 1,637 | 176 pts |
Which AI lab leads?
The lab with the most models at the top of the board is the one with the deepest bench, not only the single best model.
Anthropic holds 5 of the top 10 models (50%).
Models in the top 10 by lab
Show data table
| Lab | Top 10 |
|---|---|
| Anthropic | 5 |
| OpenAI | 3 |
| Alibaba (Qwen) | 1 |
| 1 |
Open vs proprietary: how big is the gap?
The best open-weight model is kimi-k3-max by Moonshot AI at 1,654, 159 points behind the best proprietary model, claude-opus-5.5-max by Anthropic at 1,813. Their 95% confidence intervals do not overlap, so the gap is statistically meaningful.
13 proprietary models rank above the best open-weight model, and 5 of the top 20 are open-weight.
Open-weight here means any license other than Proprietary, including licenses with usage limits. It is a broader definition than the OSI one.
Other leaderboards
LMArena runs separate arenas for different jobs. The best model for one is often not the best for another.
Other arenas
- Text (overall)
- Vision: claude-fable-5-high by Anthropic, 1,308
- Search: gpt-5.6-sol-xhigh by OpenAI, 1,256
- Document: claude-fable-5 by Anthropic, 1,499
- Text to image: gpt-image-2.5-sunburst by OpenAI, 1,425
- Image editing: gpt-image-2.5-sunburst by OpenAI, 1,524
- Text to video: gemini-omni-1.1-flash by Google, 1,516
Text arena by category
How to read this leaderboard
WebDev (Code Arena) asks models to build working web apps from a prompt. People compare the running results side by side and vote, so it rewards code that works and looks right, not code that only reads well.
- Scores come from blind votes
- LMArena shows a person two anonymous answers to the same prompt and asks which is better. Those head-to-head votes are fitted with a Bradley-Terry model, the statistical successor to Elo, to give each model one arena score. On the Elo scale LMArena uses, a 100-point lead means the higher model is expected to win about 64% of matchups.
- Confidence intervals decide ties
- Every score carries a 95% confidence interval. When two intervals overlap, the order of those two models is not settled. LMArena's rank column already accounts for this, which is why more than one model can hold the same rank.
- Why it differs from usage share
- A high rank does not mean many people use a model. Usage depends on price, distribution and habit. For adoption rather than quality, see AI market share.
Methodology and source
All rankings are read from the LMArena leaderboard dataset on Hugging Face, published by LMArena (formerly Chatbot Arena) under the Creative Commons Attribution 4.0 license. Credit for the underlying votes, scores and confidence intervals belongs to LMArena. Essentras is not affiliated with LMArena.
This page reads the webdev subset, category overall, from the dataset's latest split. Scores, intervals, ranks, vote counts and license labels are shown as published. Essentras adds the grouping by lab and the derived figures on this page: gaps, lab shares, open vs proprietary and, on the main page, the history of who held #1.
Data refreshes automatically every 2 days. If the source cannot be reached, the page falls back to the last saved copy and says so at the top. Models are never merged or renamed; model and lab names appear exactly as LMArena publishes them.
Open-weight means any license other than Proprietary. History is limited to what the dataset records: the style-controlled text leaderboard from August 28, 2024. We do not scrape any other ranking site.
Track how AI models rank and describe your brand with our AI SEO service. Want these models to name your brand? Read our guide to LLM SEO.
Frequently asked questions
What is the best AI model for web development?
As of October 8, 2026, claude-opus-5.5-max by Anthropic ranks first on LMArena's WebDev arena with a score of 1,813. Rankings move as models and votes are added, so check the table above for the current order.
Which AI lab leads for web development?
Anthropic holds 5 of the top 10 models (50%), ahead of OpenAI (3), Alibaba (Qwen) (1), and Google (1).
Is the best open-source model close to the best proprietary model?
The best open-weight model is kimi-k3-max by Moonshot AI at 1,654, 159 points behind the best proprietary model, claude-opus-5.5-max by Anthropic at 1,813. 13 proprietary models rank above the best open-weight model, and 5 of the top 20 are open-weight. Their 95% confidence intervals do not overlap, so the gap is statistically meaningful. "Open" here means any license other than Proprietary, which is broader than the OSI definition.
What does this leaderboard measure?
WebDev (Code Arena) asks models to build working web apps from a prompt. People compare the running results side by side and vote, so it rewards code that works and looks right, not code that only reads well.
How often does this leaderboard update?
This page re-reads LMArena's dataset every 2 days. LMArena publishes a new snapshot when it has gathered enough votes or added models, so the date at the top shows the latest published data.
Models rank themselves. What do they say about you?
A leaderboard shows which AI models are strongest. It does not show what those models tell your customers about your company. Run the free check and see the answers.