Skip to content

Add AI/ML API (aimlapi.com) as a built-in provider - #2

Merged
Lookoff-AIMLAPI merged 284 commits into
mainfrom
feat-aimlapi-provider
Aug 25, 2026
Merged

Add AI/ML API (aimlapi.com) as a built-in provider#2
Lookoff-AIMLAPI merged 284 commits into
mainfrom
feat-aimlapi-provider

Conversation

@Lookoff-AIMLAPI

Copy link
Copy Markdown
Member

Summary

  • Adds aimlapi as a built-in provider, following the same pattern as openrouter/vercel-ai-gateway (OpenAI-completions compatible, AIMLAPI_API_KEY).
  • Model catalog is fetched live in generate-models.ts from api.aimlapi.com/v1/models?include=pricing (353 chat models at time of writing), mapping AI/ML API's per-unit pricing schema into pi's ModelCost (including tiered/threshold pricing and a reasoning-phase signal used to set the reasoning flag).
  • Referral/attribution headers (X-AIMLAPI-Partner-ID, X-AIMLAPI-Source) are added in provider-attribution.ts, gated behind install telemetry the same way OpenRouter's own attribution headers already are.
  • Registered first (alphabetically) in all.ts/models.generated.ts, so it also surfaces first in the interactive provider picker.

Test plan

  • npm run generate-models --strict (real run, live network) completed cleanly and picked up aimlapi: 353 models alongside every other provider, with no unrelated diff to other providers' generated shards.
  • Full monorepo npm run check (biome, ts-imports, shrinkwrap, tsgo --noEmit, browser-smoke) passes via the pre-commit hook.

vegarsti and others added 30 commits August 7, 2026 10:18
Return typed results from JSONL parsers, add file context during replay, and classify torn tails explicitly.
This is incremental work towards I2 in the harness-v2.md plan.
Part of I2, "passive events and watch buffering," in packages/agent/docs/harness-v2.md.
paintBox ran every visible text row through compositeTuiLine on every
frame, rebuilding each row string via ANSI/grapheme segmentation even
when nothing changed. Full-width boxes painting onto untouched rows now
assign the source line reference directly; padding is unnecessary since
rows are written with erase-line and the final width clamp still
truncates over-wide lines.

Cuts alt-screen per-frame allocation churn 9-18x (test/render-churn-bench.ts)
and removes the fullscreen RSS penalty over regular mode on short
sessions (previously +11-29 MiB loaded, +26-36 MiB peak).
npm 12 returns an object keyed by package name instead of an array,
which broke packPackage with a TypeError.
Auto-detect DeepSeek in generated and runtime compatibility metadata so built-in and custom models use the supported token limit field.
mitsuhiko and others added 28 commits August 23, 2026 00:24
…#6847)

Add missing ToolExecutionEndEvent, ToolExecutionStartEvent, and
ToolExecutionUpdateEvent type exports to packages/coding-agent/src/index.ts
so they are available to consumers of the package.

Fixes earendil-works#6687
Registers "aimlapi" alongside the other built-in providers: a live-catalog
fetcher in generate-models.ts (api.aimlapi.com/v1/models?include=pricing
already returns per-model pricing, including tiered/threshold pricing and a
reasoning-phase signal used to set the `reasoning` flag), the generated
provider shard/catalog wiring, the aimlapi.ts provider factory
(OpenAI-completions compatible, AIMLAPI_API_KEY), and referral/attribution
headers (X-AIMLAPI-Partner-ID, X-AIMLAPI-Source) gated the same way
OpenRouter's own attribution headers already are, behind install telemetry.
@Lookoff-AIMLAPI
Lookoff-AIMLAPI merged commit 6ef842a into main Aug 25, 2026
4 checks passed
@Lookoff-AIMLAPI
Lookoff-AIMLAPI deleted the feat-aimlapi-provider branch August 25, 2026 10:58
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.