Claude Code
Anthropic's terminal coding agent. The highest-adoption CLI agent by developer survey share.
- STATE
- ACTIVE
- VENDOR
- Anthropic
- STARS
- 148.0K
- MAINTENANCE
- 94/100
- PRICING READ
- 25 Sep 2026
PLANS AND PRICESnone published as tiers
Claude Code charges on a subscription basis, or its tiers are drawn by JavaScript the reader does not run. Either way nothing on Anthropic’s page could be read as a named tier with a price beside it.
WHAT IT IS BUILT ONread off the vendor’s own pages and the repository
| DEFAULT MODEL | anthropic/claude-opus-4.8Bundled with the Claude subscription; also usable with an API key or through a gateway. |
|---|---|
| YOUR OWN KEY | Yes, you can point it at your own provider account and pay for tokens directly. |
| SOURCE | Proprietary. The client is not published. |
| HOW IT CHARGES | Subscriptionread from Anthropic’s own pricing page |
| MAINTENANCE | 94/100last pushed today |
| OPENROUTER VOLUME | 7.00Trank 4 of all apps on OpenRouter, coding and otherwise, over the trailing week |
| TAGGED | terminal, agent, anthropic, subscription |
BEST MODEL IN CLAUDE CODEcaptured 25 September 2026 · ordered by each model’s share of the tokens Claude Code sends through OpenRouter, with the bundled default first where there is one
| RANK | MODEL | SCORE | PRICE / M | CONTEXT | VOLUME |
|---|---|---|---|---|---|
| 1 |  Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 | 100.0 | $10.00 | 1M | 332B |
| 2 |  Z.ai: GLM 5.3 Flash (batch)z-ai/glm-5.3-flash:batch | 43.1 | $0.095 | 1.0M | 8.69T |
| 3 | Z.ai: GLM 5.3 (batch)z-ai/glm-5.3:batch | 8.8 | $0.84 | 1.0M | 1.77T |
| 4 |  DeepSeek: DeepSeek V4.1 Flash (batch)deepseek/deepseek-v4.1-flash:batch | 6.7 | $0.17 | 1.0M | 1.35T |
| 5 | Anthropic: Claude Opus 5 (batch)anthropic/claude-opus-5:batch | 6.1 | $5.00 | 1M | 1.22T |
| 6 |  Tencent: Hy4 previewtencent/hy4-preview | 5.3 | $1.25 | 1.0M | 1.07T |
| 7 | Anthropic: Claude Sonnet 5 (batch)anthropic/claude-sonnet-5:batch | 4.3 | $2.00 | 1M | 878B |
| 8 | Z.ai: GLM 5.2 (free)z-ai/glm-5.2:free | 4.3 | Free | 33K | 873B |
| 9 |  MiniMax: MiniMax M3minimax/minimax-m3 | 3.8 | $0.53 | 1.0M | 768B |
| 10 |  Google: Gemini 3.8 Flash (batch)google/gemini-3.8-flash:batch | 3.5 | $0.75 | 1.0M | 714B |
Rank 1 is placed first because it is the bundled default where a tool has one, not because it was measured highest. This board on the console.
WHAT PEOPLE ACTUALLY RUN IN IT20 models · measured on bring-your-own-key traffic routed through OpenRouter, a large sample of Claude Code’s users and not all of them
| RANK | MODEL | SHARE OF ITS TRAFFIC | TOKENS |
|---|---|---|---|
| 1 | z-ai/glm-5.3-flash:batch | 43.1% | 8.69T |
| 2 | z-ai/glm-5.3:batch | 8.8% | 1.77T |
| 3 | deepseek/deepseek-v4.1-flash:batch | 6.7% | 1.35T |
| 4 | anthropic/claude-opus-5:batch | 6.1% | 1.22T |
| 5 | tencent/hy4-preview | 5.3% | 1.07T |
| 6 | anthropic/claude-sonnet-5:batch | 4.4% | 878B |
| 7 | z-ai/glm-5.2:free | 4.3% | 873B |
| 8 | minimax/minimax-m3 | 3.8% | 768B |
| 9 | google/gemini-3.8-flash:batch | 3.5% | 714B |
| 10 |  stealth/ox-alpha | 2.1% | 430B |
EVERYTHING THAT HAS CHANGED6 entries since tracking began