models.dev deprecated several Baseten, GitHub Copilot, and Together models,
which broke type checks and the default-model test.
- Default GitHub Copilot model: gpt-5.4 -> gpt-6.1-sol
- Move Copilot live tests to gpt-6-luna/gpt-6-sol
- Remove tests and generator entries for retired models, including the
Together reasoning-only model list
Extensions importing the public /models subpath (e.g. clampThinkingLevel)
failed with "Cannot find module" because the loader aliased/virtualized
only the pi-ai root, /compat, /oauth, and /providers/all entrypoints.
Add /models to the jiti aliases and the embedded virtual module map,
plus a regression test importing it without a local pi-ai install.
Closes#11156
Co-authored-by: Dante-dan <dantedan@users.noreply.github.com>
Scripts wait in arrival order for a free worker instead of failing
immediately. The wait counts toward timeout_ms, honours abort, and is
capped at 60 seconds. PI_CODEMODE_MAX_WORKERS raises the process-wide
limit of two workers for hosts running many sessions in one process.
closes#11184
A second select/confirm/input/editor dialog or inline custom() component
replaced the visible one and left its promise pending forever. Dialogs in
the editor slot now form a stack: closing the visible dialog shows the
previous one again, and dialogs opened from inside a custom component work.
closes#11166
The openai-codex provider authenticates with a ChatGPT account, and the
Codex backend rejects gpt-5.5 and gpt-5.3-codex-spark for those accounts
("not supported when using Codex with a ChatGPT account"). Remove them
from the catalog so model pickers stop offering models that always fail.
Codex e2e tests that used openai-codex/gpt-5.5 now use gpt-6.1-sol, the
openai-codex default. The context overflow test uses gpt-5.6-sol to keep
its prompt smaller.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Windows PowerShell 5.1 adds a BOM when it rewrites managed-install.json, which made pi update fail with "Managed install marker is missing or invalid".
Copilot rejects request bodies over its byte-size limit with HTTP 413 and a generic
'failed to parse request' body, so the existing token-count heuristics never matched.
Treat any HTTP 413 as request-size overflow regardless of provider or body text, so
the agent compacts and retries instead of surfacing a raw API error.
Validated against the live Copilot API (40MB image payload): main classifies the 413
as a plain error, this branch detects it as overflow.
Node's existsSync() can reject runnable Store aliases because its extra
Windows-only stat check fails with EACCES (nodejs/node#36790). Use
access(F_OK) for shell validation on all platforms instead, matching the
check harness-v2's async refactor settled on (617d8b317, now pathExists
in packages/durable/src/env/node.ts). On Unix this is a no-op: existsSync
is already access(F_OK) there, so shell search order and behavior stay
unchanged; on Windows runnable aliases now pass validation.
* 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>
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
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
* 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
With an API key, GPT-6 models on the openai provider now keep reasoning.effort
fixed and send the thinking level as configuration_update input items. Each
reply records its level, so earlier items are rebuilt unchanged on every request.
Controlled by compat.supportsMidConvoEffort; skipped for Sign in with ChatGPT
and thinking off.
closes#10853
The SDK always passed initial tools, so the constructor never restored the
loadout recorded in the transcript, and resumed sessions dropped tools activated
during the session. Sessions that use the default tools now restore the recorded
loadout, with --tools +name/-name entries applied to it.
fixes#10785
Sends copied text to the terminal with OSC 52 after the local clipboard
write, as over SSH, for remote clients without SSH variables such as
multiplexer attaches and remote IDE daemons.
closes#10826, closes#10835, closes#10114
A misplaced options line was ignored, so its timeout_ms never applied and
the script ran without a deadline. parseCodemodeSource now throws
CodemodeSourceError before the script runs.
fixes#10631
A just-removed auth.json.lock directory can stay delete-pending on Windows
while another process holds a handle, making mkdir fail with EPERM. Retry
this briefly (capped separately from ELOCKED) so real permission errors
still fail fast.
fixes#10801
Print mode now writes a warning to stderr and exits 1 when the final
response stopped at the model's output token limit. The partial text
is still written to stdout.
fixes#10799
The codemode row shows "Elapsed" while the script runs and "Took" once
it finishes, like bash rows. The timer logic moves from the bash renderer
to render-utils, and scripts send an initial update so the timer starts
before the first nested call.
closes#10708
MCP 2026-07-28 allows structuredContent to be any JSON value, including
arrays and null. callTool() rejected non-object values, failing the whole
tool call. CallToolResult.structuredContent is now unknown, and MCP tools
without an output schema declare structuredContent as unknown to codemode.
refs #10416
Servers with resources registered no tools until resources/list and
resources/templates/list finished, so a slow listing (a large drive)
kept their tools out of the first prompt. The lists only fed counts in
/mcp and pi mcp list; both now show that a server offers resources.
The resource tools still list and read resources on demand.
fixes#10526
Entries like !~/skills/pdf, +~/x and -~/x in the extensions, skills,
prompts, and themes settings were compared with a literal ~ and never
matched.
fixes#10765
