[PR #321] add Open Science integration guide #342

New issue

Open

opened 2026-10-26 09:48:15 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 09:48:15 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-agent/pull/321
Author: @FerryLF
Created: 7/29/2026
Status: 🔄 Open

Base: main ← Head: add-openscience-guide

📝 Commits (5)

67f8a17 add OpenScience English and Chinese guides
2d53e51 Update openscience.zh-CN.md
2a92b70 Update openscience.md
a5dd529 Update README.md
a35a9d9 Update README.zh-CN.md

📊 Changes

4 files changed (+229 additions, -0 deletions)

View changed files

📝 README.md (+1 -0)
📝 README.zh-CN.md (+1 -0)
➕ docs/openscience.md (+114 -0)
➕ docs/openscience.zh-CN.md (+113 -0)

📄 Description

Summary

add English and Simplified Chinese guides for using DeepSeek V4 in Open Science
document installation, first-launch setup, the built-in DeepSeek provider, model selection, Max reasoning effort, and the first project run
add Open Science to both README tables in alphabetical order
add screenshots sourced from the Open Science v0.7.3 UI and onboarding manual

Compatibility and verification

checked against Open Science v0.7.3
uses only the current model names:
- deepseek-v4-pro
- deepseek-v4-pro[1m]
- deepseek-v4-flash
documents the 1,000,000-token context window declared by the v0.7.3 model catalog
confirms the built-in DeepSeek endpoints:
- https://api.deepseek.com/anthropic
- https://api.deepseek.com/v1
confirms that the UI Max reasoning selection is normalized to reasoningEffort: "max" with thinking enabled
confirms that the OpenAI-compatible bridge serializes the value into the request body as reasoning_effort: "max"
includes UI evidence for the built-in DeepSeek provider, all three current model IDs, a successful provider status, the Max reasoning option, and 71k / 1M tokens (7%) context reporting
uses the current V4 model IDs throughout, with no legacy DeepSeek model IDs
does not include pricing

Files

docs/openscience.md
docs/openscience.zh-CN.md
docs/assets/aipoch_open_science_release.png
docs/assets/aipoch_open_science_environment.png
docs/assets/aipoch_open_science_agent_runtime.png
docs/assets/aipoch_open_science_settings.jpg
docs/assets/aipoch_open_science_add_provider.jpg
docs/assets/aipoch_open_science_select_deepseek.jpg
docs/assets/aipoch_open_science_supported_models.jpg
docs/assets/aipoch_open_science_active_deepseek.jpg
docs/assets/aipoch_open_science_reasoning_effort.png
docs/assets/aipoch_open_science_context_1m.png
docs/assets/aipoch_open_science_deepseek_project.jpg
README.md
README.zh-CN.md

Verification basis

The DeepSeek integration details were checked against the exact Open Science v0.7.3 release source. The reasoning transport maps max to the model option { reasoningEffort: "max", thinking: { type: "enabled" } }, and the request-capture test for the OpenAI-compatible bridge asserts an outgoing reasoning_effort: "max" field.



---

<sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/awesome-deepseek-agent/pull/321 **Author:** [@FerryLF](https://github.com/FerryLF) **Created:** 7/29/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `add-openscience-guide` --- ### 📝 Commits (5) - [`67f8a17`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/67f8a173a961bcb50db98bf7bae77153c38a3bb0) add OpenScience English and Chinese guides - [`2d53e51`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/2d53e51d96e8e54d07b551810bf18fe4d92240d2) Update openscience.zh-CN.md - [`2a92b70`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/2a92b707ab411507d65f1bb900654f959a5e523b) Update openscience.md - [`a5dd529`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/a5dd529c21a0bba834b7947eedd01dab4f4e9098) Update README.md - [`a35a9d9`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/a35a9d9316a94c940632e56afc3d1a391d05352a) Update README.zh-CN.md ### 📊 Changes **4 files changed** (+229 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -0) 📝 `README.zh-CN.md` (+1 -0) ➕ `docs/openscience.md` (+114 -0) ➕ `docs/openscience.zh-CN.md` (+113 -0) </details> ### 📄 Description ## Summary - add English and Simplified Chinese guides for using DeepSeek V4 in Open Science - document installation, first-launch setup, the built-in DeepSeek provider, model selection, Max reasoning effort, and the first project run - add Open Science to both README tables in alphabetical order - add screenshots sourced from the Open Science v0.7.3 UI and onboarding manual ## Compatibility and verification - checked against Open Science v0.7.3 - uses only the current model names: - `deepseek-v4-pro` - `deepseek-v4-pro[1m]` - `deepseek-v4-flash` - documents the 1,000,000-token context window declared by the v0.7.3 model catalog - confirms the built-in DeepSeek endpoints: - `https://api.deepseek.com/anthropic` - `https://api.deepseek.com/v1` - confirms that the UI `Max` reasoning selection is normalized to `reasoningEffort: "max"` with thinking enabled - confirms that the OpenAI-compatible bridge serializes the value into the request body as `reasoning_effort: "max"` - includes UI evidence for the built-in DeepSeek provider, all three current model IDs, a successful provider status, the Max reasoning option, and `71k / 1M tokens (7%)` context reporting - uses the current V4 model IDs throughout, with no legacy DeepSeek model IDs - does not include pricing ## Files - `docs/openscience.md` - `docs/openscience.zh-CN.md` - `docs/assets/aipoch_open_science_release.png` - `docs/assets/aipoch_open_science_environment.png` - `docs/assets/aipoch_open_science_agent_runtime.png` - `docs/assets/aipoch_open_science_settings.jpg` - `docs/assets/aipoch_open_science_add_provider.jpg` - `docs/assets/aipoch_open_science_select_deepseek.jpg` - `docs/assets/aipoch_open_science_supported_models.jpg` - `docs/assets/aipoch_open_science_active_deepseek.jpg` - `docs/assets/aipoch_open_science_reasoning_effort.png` - `docs/assets/aipoch_open_science_context_1m.png` - `docs/assets/aipoch_open_science_deepseek_project.jpg` - `README.md` - `README.zh-CN.md` ## Verification basis The DeepSeek integration details were checked against the exact Open Science v0.7.3 release source. The reasoning transport maps `max` to the model option `{ reasoningEffort: "max", thinking: { type: "enabled" } }`, and the request-capture test for the OpenAI-compatible bridge asserts an outgoing `reasoning_effort: "max"` field. ``` --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi added the

pull-request

label

2026-10-26 09:48:15 +00:00

bsakaguchi referenced this issue

2026-10-26 09:48:22 +00:00

[PR #342] [CLOSED] Add DeepDepCat integration guide (EN + zh-CN) #356