Skip to content

feat(providers): add Huancheng Public API (hcnsec) OpenAI-compatible regional provider - #6410

Merged
diegosouzapw merged 2 commits into
release/v3.8.46from
feat/port-pr-2378-hcnsec
Jul 6, 2026
Merged

diegosouzapw merged 2 commits into
release/v3.8.46from
feat/port-pr-2378-hcnsec

Conversation

@diegosouzapw

Copy link
Copy Markdown
Collaborator

Summary

  • Adds Huancheng Public API (hcnsec) — Xinjiang Huancheng Cybersecurity's public LLM platform — as an OpenAI-compatible regional provider (base https://api.hcnsec.cn/v1, free credits via daily check-ins).
  • Wired through the shared OpenAI-compatible registry with full model passthrough — no static seed, relies on the live /v1/models catalog.
  • LLM service only. The upstream PR also bundled an image-edit endpoint and a 9router-specific build-cli standalone-dir fix; neither is ported (not applicable to OmniRoute's build).

Attribution

Thanks to @UnrealAryan for the original implementation.

Changes

  • open-sse/config/providers/registry/hcnsec/index.ts — new registry entry (buildOpenAiCompatibleRegistryEntry)
  • open-sse/config/providers/index.ts — import + wire into REGISTRY
  • src/shared/constants/providers/apikey/regional.ts — provider metadata card (passthrough, hasFree)
  • tests/unit/providers-constants-split.test.ts — count guard 168 → 169
  • tests/snapshots/provider/translate-path.json — golden regenerated (hcnsec-only insertion, +23 lines)
  • tests/unit/hcnsec-provider.test.ts — new regression guard (4 tests)

Note: the count-guard bump (168→169) collides with the concurrently-open DigitalOcean provider PR (#6373), which also bumps 168→169 — whichever merges second needs the +1 rebase. Standard provider-PR drift; handled at merge.

Test plan

  • hcnsec-provider.test.ts (4/4), providers-constants-split guard (4/4), provider-translate-path-golden (3/3)
  • npm run typecheck:core
  • npm run check:provider-consistency
  • npm run check:cycles
  • lint (changed files) clean

…regional provider

Xinjiang Huancheng Cybersecurity's public LLM platform (base
https://api.hcnsec.cn/v1, free credits via daily check-ins), wired through the
shared OpenAI-compatible registry with full model passthrough. LLM service only
(the upstream image-edit endpoint and the 9router-specific build-cli change are
not ported). Count guard 168->169, golden regenerated (hcnsec-only insertion).

Co-authored-by: Aryan <aryan.tamrakar@wowinx.com>
Inspired-by: decolua/9router#2378
@gemini-code-assist

Copy link
Copy Markdown
Contributor

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@diegosouzapw
diegosouzapw merged commit 437ca48 into release/v3.8.46 Jul 6, 2026
2 of 3 checks passed
diegosouzapw added a commit that referenced this pull request Jul 6, 2026
…#6373)

add DigitalOcean AI OpenAI-compatible provider (port PR #2417). Resolved registry conflict with hcnsec #6410; count-guard 169→170. Tests green. Integrated into release/v3.8.46.
@diegosouzapw diegosouzapw mentioned this pull request Jul 7, 2026
@diegosouzapw
diegosouzapw deleted the feat/port-pr-2378-hcnsec branch July 7, 2026 19:20
HouMinXi pushed a commit to HouMinXi/OmniRoute that referenced this pull request Aug 2, 2026
…regional provider (cheaper-inference#6410)

add Huancheng Public API (hcnsec) OpenAI-compatible provider (port PR cheaper-inference#2378) (net +1/-0, tests OK). Integrated into release/v3.8.46.
HouMinXi pushed a commit to HouMinXi/OmniRoute that referenced this pull request Aug 2, 2026
…cheaper-inference#6373)

add DigitalOcean AI OpenAI-compatible provider (port PR cheaper-inference#2417). Resolved registry conflict with hcnsec cheaper-inference#6410; count-guard 169→170. Tests green. Integrated into release/v3.8.46.
muhamadgalihsaputra pushed a commit to niyatna/NiyatnaRoute that referenced this pull request Sep 27, 2026
…regional provider (cheaper-inference#6410)

add Huancheng Public API (hcnsec) OpenAI-compatible provider (port PR cheaper-inference#2378) (net +1/-0, tests OK). Integrated into release/v3.8.46.
muhamadgalihsaputra pushed a commit to niyatna/NiyatnaRoute that referenced this pull request Sep 27, 2026
…cheaper-inference#6373)

add DigitalOcean AI OpenAI-compatible provider (port PR cheaper-inference#2417). Resolved registry conflict with hcnsec cheaper-inference#6410; count-guard 169→170. Tests green. Integrated into release/v3.8.46.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant