* docs(ai,coding-agent): generate models.json reference from schemas
Add npm run generate:references, which renders docs/models-configuration.md
from the models.json TypeBox schemas, and describe provider and model fields
in those schemas. Overrides get their own descriptions without defaults, and
compat sections cover the Responses API aliases.
closes#10790
* Modify changelog
* docs(ai,coding-agent): use shared thinkingLevelMap and promptCache schemas
Replace the per-use schema helpers with fixed schemas whose descriptions
hold for both models and modelOverrides entries.
* feat(coding-agent): default NVIDIA to Nemotron 3.5 Super VL
Co-authored-by: prademacher-nvidia <235594174+prademacher-nvidia@users.noreply.github.com>
* docs(coding-agent): changelog for NVIDIA default model change
* fix(ai): run NVIDIA E2E tests against Nemotron 3.5 Super VL
---------
Co-authored-by: prademacher-nvidia <235594174+prademacher-nvidia@users.noreply.github.com>
The fake tool API lacked retainedOutput, so every bash case threw on both sides
and the test only compared the error text. Its stdout+stderr case also depended
on the order of two concurrently read pipes and failed intermittently in CI.
The API now provides retainedOutput, each bash case writes to one stream, and
bash cases must return a result.
Codex no longer serves gpt-5.5 to ChatGPT accounts ("The 'gpt-5.5' model is not
supported when using Codex with a ChatGPT account"), so the test failed in its
first turn as unanswered.
Like env, models and settings may be a function returning the current value.
Generation, compaction, hooks and tools read it at each use (api.models at each
access), so a host can replace its model configuration without reopening the
Harness. A plain object still works.
Whitespace-only tool results (e.g. a bare \r\n from bash) were sent verbatim
to OpenAI Completions, OpenAI Responses, and Google providers. Some providers
reject that with a 400, which then repeats on every request in the session.
They now use the same placeholder as empty output.
closes#8720
Compaction summaries were limited to 0.8 * compaction.reserveTokens (13107 by default), turn-prefix summaries to 0.5 * reserveTokens, and branch summaries to a hardcoded 4096. On providers where reasoning shares the output limit (OpenAI Responses, adaptive Anthropic), high thinking levels used most of that limit and compaction failed with "generation hit the token cap", even though the model allows far more output.
Tying the limit to reserveTokens assumed the summary request is as large as the full context, so its output had to fit in the reserved space. It is usually much smaller: recent messages kept by keepRecentTokens are not sent, a split turn goes to a separate request, tool results are truncated, and tool definitions and the normal system prompt are not included. Manual compaction can also run long before the context is full. The provider layer already sizes output from the actual request, using the model's output limit reduced to fit the remaining context, as for normal turns.
Summary requests now set no maxTokens and rely on that. compaction.reserveTokens only controls when compaction starts.
generateSummary() and generateSummaryWithUsage() now ignore the reserveTokens parameter.
closes#9512, closes#9075, closes#10908
pi.dev serves each published catalog to released pi >= MINIMUM_PI_VERSION.
Compile the latest release's pi-ai against its built-in models overlaid by
the new catalog, so a catalog it cannot handle (e.g. kimi-coding moving to
openai-completions) fails in CI instead of breaking released clients.
MINIMUM_PI_VERSION moves to scripts/model-catalog-minimum-version.mjs.
done() from ctx.ui.custom({ overlay: true }) closed the overlay with
ui.hideOverlay(), which removes the topmost overlay instead of the one
that called done(). With a shell overlay open under a dialog overlay
from another extension, the shell's exit removed the dialog: the
finished shell overlay stayed on screen ignoring input and the dialog's
promise never resolved. Hide the overlay through its own handle.
closes#10648
The pin script discovered the catalog with the checkout's version, which on
main is the last release, so pi.dev answered with the catalog for released
clients after MINIMUM_PI_VERSION was raised. It now pins the newest
published catalog and treats the pin as stale when its provider/type/API
model groups differ. The release script pins the newest catalog and runs
the Nix build steps before tagging.
closes#10985
When switching models, thinking blocks from earlier assistant messages
produced by a different model were sent as plain text. The new model saw
the old model's reasoning as part of its replies and could start writing
its own reasoning into the visible answer. These blocks are now dropped;
text and tool calls are kept.
closes#10995
Anthropic and Claude on Bedrock reject error tool results that contain
images. Instead of replacing the images with a note, each image in the
result is now replaced with a placeholder and the images are sent after
all tool results of the turn, in the same user message, under a caption
naming the tool call. Successful results, error results without images,
and non-Claude Bedrock models are unchanged.
closes#10794
Switch kimi-coding to OpenAI Chat Completions at api.kimi.com/coding/v1,
the protocol the official Kimi Code client uses. K3, K3-256K, and Kimi For
Coding accept tools declared in later system messages, so tools loaded
mid-conversation (e.g. by tool_search) keep the prompt cache. Highspeed
rejects tool-bearing system messages and keeps added tools top-level.
Re-lands b06c1f2 (reverted in 6a1f799b6) with MINIMUM_PI_VERSION bumped
to 1.2.1. Released clients implement only Anthropic Messages for
kimi-coding and ignore a model's api, so they must keep receiving the
last catalog published under 0.80.7.
closes#10720
Reverts b06c1f2. The pi.dev model catalog is served to every client
since 0.80.7, so released clients received kimi-coding models with
api openai-completions and baseUrl https://api.kimi.com/coding/v1.
Their kimi-coding provider only implements Anthropic Messages and
ignores model.api, so requests went to /coding/v1/v1/messages and
failed with 404. The change will re-land together with a bump of the
catalog's minimum pi version.
refs #10720, refs #10985
OutputBuffer, Progress, boundOutput, sanitizeOutput, characterEnd and the
output types move from Durable's internal harness/output.ts to the new
Chord subpath, with OutputSkip (the shape of Durable's ShellOutputSkip) and
utf8ByteLength. Durable imports them from Chord; its truncate.ts re-exports
Chord's utf8ByteLength instead of keeping a second copy.
Both models accept thinking: {type: "disabled"}, but were treated like
Opus 5.5, which cannot turn thinking off. "off" now sends disabled
thinking without effort markers or block_binding, which Anthropic
rejects while thinking is disabled.
fixes#10753
A call through a remote reference checked the ctx against its own host,
so a ctx from another host, or an object not made by Chord, passed and
its root was carried over the link. It now runs the same checks as a
local reference and rejects with invalid/foreign_ctx.
closes#10959
Switch kimi-coding to OpenAI Chat Completions at api.kimi.com/coding/v1,
the protocol the official Kimi Code client uses. K3, K3-256K, and Kimi For
Coding accept tools declared in later system messages, so tools loaded
mid-conversation (e.g. by tool_search) keep the prompt cache. Highspeed
rejects tool-bearing system messages and keeps added tools top-level.
closes#10720
When a finished tool call's arguments text was not valid JSON, the
streaming parser kept only the fields before the error and the tool ran
with them. Finished calls are now parsed strictly (with the existing
string repair); on failure arguments are {} and argumentsError holds the
reason and raw text, which validateToolArguments reports to the model.
fixes#10897
- an awaited get or a ready hold has every state member of its variant, local or remote
- holds subscribe to their variant's state; per-member opt-out with state: false
- remote binding keys use canonical identity; client holds carry an opaque origin
- shape frames no longer carry codeVersion
- self.state(name, schema, { init }) loads before the object goes live
- the principal type is a parameter; createHost self, connect and route principals are required
- remote contracts declare their host (hostNamed, hostFamily); get no longer takes at
MiMo Chat Completions rejects reasoning_effort "minimal" with a 400.
Xiaomi documents only thinking.type enabled/disabled and treats all effort
levels identically, so stop sending reasoning_effort and expose only the
off and high thinking levels.
fixes#10818
Proxies such as LiteLLM can send reasoning_content and content in one
stream delta. Reasoning was handled after content, so the message ended
up as [text, thinking]. Handle reasoning fields and reasoning_details
before content and tool calls.
fixes#10926
* fix(coding-agent): resolve model references through one provider selection policy
Model references from --model, --models/enabledModels, /model, and auth credential printing now share one resolver. It collects every candidate before choosing a provider, so catalog order never decides, a :<thinking> suffix cannot change the provider, and an ambiguous exact match no longer falls through to a fuzzy match.
provider/id still prefers the named provider's exact or fuzzy match. A gateway exposing provider/id as a raw model id is used when the named provider has no match, or lacks credentials while the gateway has them. Several gateways without a unique authenticated one are reported as ambiguous.
fixes#10160
* docs(coding-agent): shorten model selection notes in CLI reference
* docs(coding-agent): condense #10160 changelog entry
* fix(coding-agent): move model reference resolution into model-resolver
* fix(coding-agent): warn when a model scope pattern matches several providers
* docs(coding-agent): trim model resolver comments to the reasons
* fix(coding-agent): build custom model ids from the resolver's not-found target
Context derivation moved a system message to the front only when user messages
alone preceded it. After a compaction that keeps an assistant message, the
baseline written for the next run followed the kept history and the new input,
so Anthropic and OpenAI Codex requests had no initial prompt or tool set.
Derivation now identifies the harness baseline: the first pi.system entry after
the head marker, or of the transcript when no earlier entry carries a system
message, even one omitted later. Its message leads even past assistant
messages. Only the first system message ever moves, so replay is unchanged;
other system messages after an assistant message stay in place.
closes#10904
serve(Type, factory, { lifetime: "host" }) compiles only for a local type whose input has no required fields, and
not together with linger; the runtime check stays as a backstop. ServeOptions is a union (ReferencedServeOptions,
HostServeOptions), picked per contract by ServeOptionsFor<C>.
toBase64 grew each 8 KiB chunk a character at a time; V8 kept every
concatenation node until the final join, about 138 MB of heap for a
3 MiB image, beyond a Durable Object's 128 MB limit.
closes#10923
