chore: sync models with dashboard API

[amaan-ai20]

Automated model-catalog sync. The dashboard API is the source of truth; every value below was taken from it. Merging publishes v3.10.0 to npm.

Added

deepseek-v4.1-flash — Text Generation, maxTokens 1,048,576, $0.30 in / $1.20 out per 1M tokens.

Surface	Change
src/lib/savings.ts	ZGPU_PRICING entry { in: 0.3, out: 1.2 }
tests/savings.test.ts	matching CATALOG entry
src/commands/chat.ts	CHAT_MODELS entry, routed to Responses
README.md	row in the chat Models table
docs/DOCUMENTATION.md	row in the chat Models table (API column Responses) and the §1 "Reasoning and tool-use chat" bullet

The API payload carries a sample_responses_body for this model and no parameters value, so:

• it is routed to Responses, per the skill's endpoint rule for a new chat model. The route can be widened later if an explicit confirmation says otherwise.
• the notes cell describes the architecture from pricing.description ("sparse mixture-of-experts … activating 8B parameters on input and 16B on output") rather than stating a total parameter count, since the API reports none.

ZGPU_FALLBACK is unchanged: glm-5.2 still holds both the highest input and the highest output rate.

Nothing else drifted — no corrections, renames, or removals this run. The remaining models the API prices with no CLI command are unchanged: zlm-v1-moderation-edge, zlm-v1-signal-extract, all-minilm-l6-v2, bge-small-en-v1.5.

Verification

• audit-models.py --strict — 0 findings, clean apart from NOTE lines
• npm run lint, npm run build, npm test (56 tests) pass
• zerogpu chat --help lists exactly the API's 8 Text Generation models

🤖 Generated with Claude Code

https://claude.ai/code/session_01PX3a2wyDoH8pjweMjZuANg

---

Generated by Claude Code

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: 24fcbb56-cfec-4313-8c36-6196803b1522

---

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