# AI Coding Model Board

> As of 2026-10-14, among models on this board with a published CursorBench 4.0 score, Claude Opus 5.5 leads at 57.8%, and Gemini 3.8 Flash has the lowest listed standard input price at $0.75 per 1M tokens.

> 截至 2026-10-14，本表已公布 CursorBench 4.0 分数的模型中，Claude Opus 5.5 最高，为 57.8%；已列出的标准输入价里，Gemini 3.8 Flash 最低，为每百万 token $0.75。

Public list prices per 1M tokens, context windows, and published SWE-bench Verified and CursorBench scores. Blank fields are unpublished, not estimates.

公开标价（每百万 token）、上下文长度，以及已公布的 SWE-bench Verified 与 CursorBench 分数。空字段表示来源未公布，不是估算。

- Updated: 2026-10-14
- License: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). This compilation is CC BY 4.0. The prices and scores remain the publishers'. 本汇编采用 CC BY 4.0。价格与分数仍归各发布方。
- Creator: [Evan Long](https://evanlong.me) ([Follow @evanlong_me on X](https://x.com/evanlong_me))
- [English page](https://evanlong.me/en/models)
- [中文页面](https://evanlong.me/zh/models)
- [JSON](https://evanlong.me/models.json)

## Models

### Claude Opus 5.5

- Provider: Anthropic
- API id: claude-opus-5-5
- Input: $4 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Output: $20 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Context: 1M ([source](https://platform.claude.com/docs/en/models/overview), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 57.8% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - CursorBench 4.0, Max effort.
  - 中文：CursorBench 4.0，Max effort。

### Claude Sonnet 5.5

- Provider: Anthropic
- API id: claude-sonnet-5-5
- Input: $2 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Output: $10 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Context: 1M ([source](https://platform.claude.com/docs/en/models/overview), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 55.5% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - CursorBench 4.0, Max effort.
  - 中文：CursorBench 4.0，Max effort。

### Gemini 4 Argon

- Provider: Google
- API id: gemini-4-argon
- Input: $2 per 1M tokens ([source](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-14)
- Output: $10 per 1M tokens ([source](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-14)
- Context: 1M ([source](https://ai.google.dev/gemini-api/docs/models), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 52.9% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - Standard paid tier through November 30, 2026. The listed rate becomes $4 input and $20 output on December 1, 2026.
  - CursorBench 4.0, High effort.
  - 中文：标准付费档价格截至 2026-11-30。2026-12-01 起标价改为输入 $4、输出 $20。
  - 中文：CursorBench 4.0，High effort。

### Claude Fable 5.1

- Provider: Anthropic
- API id: claude-fable-5-1
- Input: $10 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Output: $50 per 1M tokens ([source](https://platform.claude.com/docs/en/about-claude/pricing), read 2026-10-14)
- Context: 1M ([source](https://platform.claude.com/docs/en/models/overview), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 51.8% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - CursorBench 4.0, Max effort.
  - 中文：CursorBench 4.0，Max effort。

### Grok 4.7

- Provider: xAI
- API id: grok-4.7
- Input: $2 per 1M tokens ([source](https://docs.x.ai/docs/models), read 2026-10-14)
- Output: $6 per 1M tokens ([source](https://docs.x.ai/docs/models), read 2026-10-14)
- Context: 500K ([source](https://docs.x.ai/docs/models), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 46.3% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - Price for prompts under 200K tokens. A prompt of 200K tokens or more is billed at $4 input and $12 output per 1M tokens for the whole request.
  - CursorBench 4.0, Extra High effort.
  - 中文：提示少于 200K token 的价格。达到 200K token 及以上时，整次请求按输入 $4、输出 $12（每百万 token）计费。
  - 中文：CursorBench 4.0，Extra High effort。

### GLM-5.3

- Provider: Z.ai
- API id: glm-5.3
- Input: $1.4 per 1M tokens ([source](https://docs.z.ai/guides/overview/pricing), read 2026-10-14)
- Output: $4.4 per 1M tokens ([source](https://docs.z.ai/guides/overview/pricing), read 2026-10-14)
- Context: 1M ([source](https://docs.z.ai/guides/llm/glm-5.3), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 42.6% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - CursorBench 4.0, Max effort.
  - 中文：CursorBench 4.0，Max effort。

### DeepSeek V4 Pro

- Provider: DeepSeek
- API id: deepseek-v4-pro
- Input: $1.32 per 1M tokens ([source](https://api-docs.deepseek.com/quick_start/pricing), read 2026-10-14)
- Output: $3.96 per 1M tokens ([source](https://api-docs.deepseek.com/quick_start/pricing), read 2026-10-14)
- Context: 1M ([source](https://api-docs.deepseek.com/quick_start/pricing), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: not published
- Notes:
  - Cache-miss price during peak hours (01:00–04:00 and 06:00–10:00 UTC, Monday–Friday, excluding Chinese public holidays). Off-peak is half. The docs list the model version as DeepSeek-V4-Pro-0813.
  - 中文：高峰时段缓存未命中价格（UTC 周一至周五 01:00–04:00 与 06:00–10:00，不含中国法定节假日）。非高峰为半价。文档中的模型版本是 DeepSeek-V4-Pro-0813。

### GPT-5.6 Sol

- Provider: OpenAI
- API id: gpt-5.6-sol
- Input: $4 per 1M tokens ([source](https://developers.openai.com/api/docs/models/gpt-5.6-sol), read 2026-10-14)
- Output: $20 per 1M tokens ([source](https://developers.openai.com/api/docs/models/gpt-5.6-sol), read 2026-10-14)
- Context: 1.05M ([source](https://developers.openai.com/api/docs/models/gpt-5.6-sol), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 41.7% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - Prompts over 272K input tokens are billed at 2× input and 1.5× output for the whole request. OpenAI lists this rate as promotional at least through November 21, 2026.
  - CursorBench 4.0, Max effort.
  - 中文：输入超过 272K token 时，整次请求按输入 2 倍、输出 1.5 倍计费。 OpenAI 称此价格至少优惠到 2026-11-21。
  - 中文：CursorBench 4.0，Max effort。

### GPT-5.6 Terra

- Provider: OpenAI
- API id: gpt-5.6-terra
- Input: $2 per 1M tokens ([source](https://developers.openai.com/api/docs/models/gpt-5.6-terra), read 2026-10-14)
- Output: $12 per 1M tokens ([source](https://developers.openai.com/api/docs/models/gpt-5.6-terra), read 2026-10-14)
- Context: 1.05M ([source](https://developers.openai.com/api/docs/models/gpt-5.6-terra), read 2026-10-14)
- SWE-bench Verified: not published
- CursorBench: 41.3% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - Prompts over 272K input tokens are billed at 2× input and 1.5× output for the whole request.
  - CursorBench 4.0, Max effort.
  - 中文：输入超过 272K token 时，整次请求按输入 2 倍、输出 1.5 倍计费。
  - 中文：CursorBench 4.0，Max effort。

### Gemini 3.8 Flash

- Provider: Google
- API id: gemini-3.8-flash
- Input: $0.75 per 1M tokens ([source](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-14)
- Output: $3.75 per 1M tokens ([source](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-14)
- Context: not published
- SWE-bench Verified: not published
- CursorBench: 39.6% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - Standard paid tier through December 31, 2026. The listed rate becomes $1.50 input and $7.50 output on January 1, 2027.
  - CursorBench 4.0, High effort.
  - 中文：标准付费档价格截至 2026-12-31。2027-01-01 起标价改为输入 $1.50、输出 $7.50。
  - 中文：CursorBench 4.0，High effort。

### Composer 2.5

- Provider: Cursor
- API id: not published
- Input: not published
- Output: not published
- Context: not published
- SWE-bench Verified: not published
- CursorBench: 27.7% ([source](https://cursor.com/cursorbench), read 2026-10-14)
- Notes:
  - CursorBench 4.0. The published row does not name an effort level.
  - 中文：CursorBench 4.0。公布的这一行没有标注 effort。

## Citations

- https://platform.claude.com/docs/en/about-claude/pricing
- https://platform.claude.com/docs/en/models/overview
- https://cursor.com/cursorbench
- https://ai.google.dev/gemini-api/docs/pricing
- https://ai.google.dev/gemini-api/docs/models
- https://docs.x.ai/docs/models
- https://docs.z.ai/guides/overview/pricing
- https://docs.z.ai/guides/llm/glm-5.3
- https://api-docs.deepseek.com/quick_start/pricing
- https://developers.openai.com/api/docs/models/gpt-5.6-sol
- https://developers.openai.com/api/docs/models/gpt-5.6-terra