Skip to content

Reconcile Nebius provider metadata - #1864

Open
DanielButler1 wants to merge 2 commits into
mainfrom
audit/nebius-20260824
Open

Reconcile Nebius provider metadata#1864
DanielButler1 wants to merge 2 commits into
mainfrom
audit/nebius-20260824

Conversation

@DanielButler1

@DanielButler1 DanielButler1 commented Aug 24, 2026

Copy link
Copy Markdown
Member

Summary

  • reconcile all 30 active Nebius Token Factory serverless deployments from the official live feed
  • retain nebius as the catalogue-only alias linked to the gateway-backed nebius-token-factory family
  • add exact case-sensitive API model IDs, canonical entities, modalities, capabilities, contexts, quantization, regions, and the standard service tier
  • add structured per-million-token pricing for every active deployment exactly as published
  • correct Qwen3-Embedding-8B from text generation to embeddings
  • preserve 10 prior catalogue aliases as disabled historical records and leave every nebius-token-factory* gateway/pricing record untouched
  • add a checked-in 30/30 reconciliation ledger with zero unaccounted deployments

Official architecture decision

Nebius AI Studio exposes these models through the Token Factory OpenAI-compatible API at api.tokenfactory.nebius.com. Phaseo's nebius-token-factory* slugs are the existing routable gateway offers and retain their history. The nebius slug is a non-routable catalogue alias, now explicitly related through provider_family_id; the live serverless catalogue is reconciled there without rewriting gateway history.

Reconciliation

  • active official deployments: 30
  • represented active deployments: 30
  • unaccounted: 0
  • historical catalogue records preserved disabled: 10
  • modalities: text generation, vision-chat, video-input chat where published, embeddings
  • execution/data regions: model-specific eu-north1, us-central1, uk-south1, or eu-west2
  • service tier: standard
  • max output: left null because the live provider feed publishes context limits but no distinct maximum-output limits
  • pricing: all published input/output token prices; no inferred cache pricing

Primary sources

Validation

  • pnpm validate:data
  • pnpm validate:pricing
  • pnpm validate:gateway
  • git diff --check

Created with Codex

@vercel

vercel Bot commented Aug 24, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
phaseo Ready Ready Preview Aug 24, 2026 3:19am

Request Review

@coderabbitai

coderabbitai Bot commented Aug 24, 2026

Copy link
Copy Markdown

Warning

Review limit reached

Next included review available in 19 minutes.

View limit details

Limit details: You’ve used the included review currently available.

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 29cad6e0-6c39-452a-9ffb-80e5fded9f89

📥 Commits

Reviewing files that changed from the base of the PR and between cc2886d and be80efa.

📒 Files selected for processing (36)
  • .changeset/nebius-provider-family-audit.md
  • packages/data/catalog/src/data/api_providers/nebius/api_provider.json
  • packages/data/catalog/src/data/api_providers/nebius/models.json
  • packages/data/catalog/src/data/api_providers/nebius/reconciliation-2026-08-24.json
  • packages/data/catalog/src/data/models/openbmb/minicpm-v-4.5/model.json
  • packages/data/catalog/src/data/organisations/openbmb/organisation.json
  • packages/data/catalog/src/data/pricing/nebius/deepseek-deepseek-v4-flash-0731/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/deepseek-deepseek-v4-flash/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/deepseek-deepseek-v4-pro/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/google-gemma-3-27b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/meta-llama-3.3-70b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/minimax-minimax-m2.5/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/minimax-minimax-m3/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/moonshotai-kimi-k2.6/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/moonshotai-kimi-k2.7-code/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/moonshotai-kimi-k3/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nous-hermes-4-405b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nous-hermes-4-70b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-cosmos3-super-reasoner/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-llama-3.1-nemotron-ultra-253b-v1/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-nemotron-3-nano-30b-a3b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-nemotron-3-nano-omni/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-nemotron-3-super-120b-a12b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-nemotron-3-ultra-550b-a55b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/nvidia-nemotron-3.5-lightning/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/openai-gpt-oss-120b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/openbmb-minicpm-v-4.5/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen2.5-vl-72b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3-235b-a22b-2507/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3-30b-a3b-instruct-2507/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3-32b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3-embedding-8b/embeddings/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3-next-80b-a3b-thinking/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/qwen-qwen3.5-397b-a17b/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/z-ai-glm-5.1/text.generate/pricing.json
  • packages/data/catalog/src/data/pricing/nebius/z-ai-glm-5.2/text.generate/pricing.json

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@github-actions

Copy link
Copy Markdown
Contributor

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 0cd7754202

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

Comment on lines +44 to +47
"kind": "official_api",
"url": "https://tokenfactory.nebius.com/api/public/models_info",
"accessed_at": "2026-08-24T00:00:00Z",
"notes": "Live public model catalogue; authoritative for active serverless model IDs, modalities, regions, context limits, and per-million-token prices."

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Add the UK region before marking region data verified

For the uk-south1 deployment identified as active by this audit, the provider still publishes default_execution_regions and default_data_regions as only EU, US, and IL (lines 10–11). Provider filters and model quickstart residency displays consume those fields, so declaring this feed authoritative for active regions while omitting the UK leaves the newly verified catalogue reporting an incomplete residency footprint; add the corresponding UK region and reconcile the residency notes, or avoid claiming region verification.

AGENTS.md reference: packages/data/catalog/AGENTS.md:L14-L16

Useful? React with 👍 / 👎.

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