[PR #464] docs(pi): document built-in DeepSeek provider and add screenshots

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/awesome-deepseek-agent/pull/464 **Author:** [@SamYue1](https://github.com/SamYue1) **Created:** 10/19/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `docs/pi-deepseek-guide` --- ### 📝 Commits (1) - [`e6c7276`](https://github.com/deepseek-ai/awesome-deepseek-agent/commit/e6c7276ab1ace6dcb1480aeb4434a98c28d1873b) docs(pi): document built-in DeepSeek provider and add screenshots ### 📊 Changes **8 files changed** (+236 additions, -186 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -1) 📝 `README.zh-CN.md` (+1 -1) ➕ `docs/assets/pi_deepseek_first_run.png` (+0 -0) ➕ `docs/assets/pi_deepseek_login.png` (+0 -0) ➕ `docs/assets/pi_deepseek_model_picker.png` (+0 -0) ➕ `docs/assets/pi_deepseek_thinking_level.png` (+0 -0) 📝 `docs/pi_mono.md` (+118 -93) 📝 `docs/pi_mono.zh-CN.md` (+116 -91) </details> ### 📄 Description ## Summary Refreshes the Pi integration guide now that Pi ships a built-in DeepSeek provider. - Pi 1.3 includes `deepseek` in its model catalog, so setup is just an API key (`/login deepseek` or `DEEPSEEK_API_KEY`) — no hand-written `models.json`, custom provider block, or compatibility proxy. - Replaces the retired `deepseek-v4-flash` with the current `deepseek-flash`, documents `deepseek-v4-pro`, 1M context, 384K output, and the `off` / `high` / `max` thinking levels. - Updates pricing to the current peak/off-peak table and notes that Pi's in-session estimate uses the peak rates. - Adds four screenshots captured from a real Pi 1.3.0 session against DeepSeek. - Updates the Pi row in both READMEs and fixes the stale `badlogic/pi-mono` links. Refs #463 ## Verification Tested on Pi 1.3.0 with a real DeepSeek key, using clean agent directories for each credential path: - `npm view` → 1.3.0 with `engines: node >=22.19.0`; the installer URL returns HTTP 200. - `/login deepseek` stores a credential in `auth.json`; a new process using only that file answers correctly. - `DEEPSEEK_API_KEY` and `--api-key` both work. - `pi auth check --provider deepseek --json` → `{"status":"ready","provider":"deepseek","authType":"api_key"}` for both stored and environment credentials. - Both model IDs return responses. Thinking levels measurably change reasoning tokens on the same prompt: `off` = 0, `high` = 6, `max` = 12. - Fresh interactive sessions default to `deepseek-v4-pro` with `high` thinking and a `1.0M` context; `/model` lists both built-in models and marks `deepseek-v4-pro` as `default`. ## Checklist ### Agent Configuration - [x] Model names are current (`deepseek-v4-pro` / `deepseek-flash`, not V3 names) - [x] 1M context is configured or mentioned - [x] Max thinking effort level is supported - [x] Pricing is current (input, output, cache hit), verified against [DeepSeek API Docs](https://api-docs.deepseek.com/quick_start/pricing) - [x] Config fields are verified to work in the agent - [x] No workarounds for already-fixed upstream bugs ### Documentation - [x] Both English and Chinese versions included in a single PR - [x] README table entry inserted in alphabetical order - [x] One tool per PR; unrelated tools not combined - [x] Images placed in `docs/assets/` with descriptive filenames and reasonable sizes --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi commented

2026-10-26 09:49:50 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/awesome-deepseek-agent/pull/464
Author: @SamYue1
Created: 10/19/2026
Status: 🔄 Open

Base: main ← Head: docs/pi-deepseek-guide

📝 Commits (1)

e6c7276 docs(pi): document built-in DeepSeek provider and add screenshots

📊 Changes

8 files changed (+236 additions, -186 deletions)

View changed files

📝 README.md (+1 -1)
📝 README.zh-CN.md (+1 -1)
➕ docs/assets/pi_deepseek_first_run.png (+0 -0)
➕ docs/assets/pi_deepseek_login.png (+0 -0)
➕ docs/assets/pi_deepseek_model_picker.png (+0 -0)
➕ docs/assets/pi_deepseek_thinking_level.png (+0 -0)
📝 docs/pi_mono.md (+118 -93)
📝 docs/pi_mono.zh-CN.md (+116 -91)

📄 Description

Summary

Refreshes the Pi integration guide now that Pi ships a built-in DeepSeek provider.

Pi 1.3 includes deepseek in its model catalog, so setup is just an API key (/login deepseek or DEEPSEEK_API_KEY) — no hand-written models.json, custom provider block, or compatibility proxy.
Replaces the retired deepseek-v4-flash with the current deepseek-flash, documents deepseek-v4-pro, 1M context, 384K output, and the off / high / max thinking levels.
Updates pricing to the current peak/off-peak table and notes that Pi's in-session estimate uses the peak rates.
Adds four screenshots captured from a real Pi 1.3.0 session against DeepSeek.
Updates the Pi row in both READMEs and fixes the stale badlogic/pi-mono links.

Refs #463

Verification

Tested on Pi 1.3.0 with a real DeepSeek key, using clean agent directories for each credential path:

npm view → 1.3.0 with engines: node >=22.19.0; the installer URL returns HTTP 200.
/login deepseek stores a credential in auth.json; a new process using only that file answers correctly.
DEEPSEEK_API_KEY and --api-key both work.
pi auth check --provider deepseek --json → {"status":"ready","provider":"deepseek","authType":"api_key"} for both stored and environment credentials.
Both model IDs return responses. Thinking levels measurably change reasoning tokens on the same prompt: off = 0, high = 6, max = 12.
Fresh interactive sessions default to deepseek-v4-pro with high thinking and a 1.0M context; /model lists both built-in models and marks deepseek-v4-pro as default.

Checklist

Agent Configuration

Model names are current (deepseek-v4-pro / deepseek-flash, not V3 names)
1M context is configured or mentioned
Max thinking effort level is supported
Pricing is current (input, output, cache hit), verified against DeepSeek API Docs
Config fields are verified to work in the agent
No workarounds for already-fixed upstream bugs

Documentation

Both English and Chinese versions included in a single PR
README table entry inserted in alphabetical order
One tool per PR; unrelated tools not combined
Images placed in docs/assets/ with descriptive filenames and reasonable sizes

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

bsakaguchi added the

pull-request

label

Rows
Columns

[PR #464] docs(pi): document built-in DeepSeek provider and add screenshots #456