MiniMax-AI / MiniMax-AI/MiniMax-MCP-JS
[Feature Request] MiniMax M2.7 模型支持图片输入 (Vision Input Support)
Open
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 130
- Forks
- 42
- PR merge metrics
- No merged PRs in 30d
Description
问题描述
MiniMax M2.7 / M2.7 Highspeed 模型不支持图片输入（`input: ["text"]`），导致在使用 OpenClaw 等 MCP 客户端时，用户无法直接发送截图——图片在网关层就被丢弃。
现状
- MiniMax M2.7 系列：`input: ["text"]`（不支持图片）
- 竞品 Kimi K2.5：`input: ["text", "image"]`（支持图片）
- MiniMax 提供了 `minimax-coding-plan-mcp` 中的 `understand_image` 工具支持图片理解，但该工具无法绕过 MCP 客户端的网关图片过滤层
期望
MiniMax M2.7 及以上模型增加图片输入支持（`input: ["text", "image"]`），与 moonshot/kimi-k2.5 等竞品看齐，提升在 Agent / MCP 场景下的用户体验。
相关链接
- OpenClaw Bug Report: https://github.com/openclaw/openclaw/issues/68513
- MiniMax M2.7 发布: https://www.minimaxi.com/news/minimax-m27-zh
环境
- Model: minimax/m2.7-highspeed
- MCP: minimax-coding-plan-mcp (含 understand_image 工具)
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no files, tests, or entry points. Start by locating where MiniMax model capabilities and input modalities are declared, then verify whether the M2.7 API accepts image input. Done means M2.7 and M2.7 Highspeed advertise image input with coverage for the changed behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- ai, api
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100