feat(ai): add Neuralwatt provider - #1051
Open
sk0x0y wants to merge 5 commits into
Open
Conversation
Add Neuralwatt (https://api.neuralwatt.com/v1, env NEURALWATT_API_KEY) as a built-in OpenAI-completions provider with a 22-model static catalog. - catalog source of truth in packages/ai/scripts/generate-models.ts - provider factory, env key mapping, KnownProvider union, builtin registration - unit + key-gated cross-provider tests - coding-agent + infra mirror updates (docs, env lists, devcontainer, test scripts)
…lwatt # Conflicts: # packages/ai/src/providers/data/.manifest.json # packages/coding-agent/src/core/changes.md
Run npm run hydrate-model-data so data/.manifest.json carries the neuralwatt.json sha256 + structureHash, per the providers AGENTS.md generation contract (data/ is committed generated source; validate with npm run check:model-data — passing).
Author
|
Synced with latest main ( For context since the original submission: Neuralwatt's model entries are now tracked in models.dev (anomalyco/models.dev#5561 — includes DeepSeek V4 Pro, Qwen 3.8 27B, V4 Flash Flex), so the generated catalog stays current with the upstream data source. I've also proposed the provider upstream (earendil-works/pi#8749) with a port of this implementation. Happy to restructure anything if there's a preferred convention for provider additions. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Add Neuralwatt (
https://api.neuralwatt.com/v1, envNEURALWATT_API_KEY) as a built-in OpenAI-completions provider with a 22-model static catalog generated frompackages/ai/scripts/generate-models.ts. Includes provider factory, registration, env-key mapping, unit + key-gated cross-provider tests, and coding-agent/infra mirror updates.Why this belongs in core
Built-in provider catalogs and env-key detection live in
packages/aicore; a per-usermodels.jsonoverlay cannot register a builtin catalog provider or contribute toKnownProvider/builtinProviders/env-api-keys. Adding Neuralwatt as a core provider lets users configure only an API key.Verification
npm run checkgreenCI=1 npm testgreen (13732 passed / 929 skipped / 0 failed)npm run check:model-datagreenCompatibility
Purely additive; no existing provider modified. The only generated manifest update is for the new
neuralwatt.jsoncatalog.Summary by cubic
Adds Neuralwatt as a built-in OpenAI-completions provider with a 22-model catalog at https://api.neuralwatt.com/v1. Users previously needed per-user overlays to reach it; now setting
NEURALWATT_API_KEYis enough.packages/aicore:KnownProvider,builtinProviders(), env-key map, and generated model data.neuralwattdefault modelkimi-k2.6-fastand display name to the coding agent.--help, devcontainer, and test-script env lists.Rollout
NEURALWATT_API_KEYto use; live tests only run when it's present.Written for commit 7994183. Summary will update on new commits.