Make strict mode as unsupported in cerebras, as mixed strict/unstrict tool usage doesn't work (400 from API), and extensions may define both strict and unstrict tools
Device authorization flow against auth.meta.com mints a Model API key via the Muse Code key-mint endpoint. Identity token stored as refresh, minted key as access; daily re-mint via the standard OAuth scheduler. Identity is non-renewable (no refresh grant), so mint 401/403 directs to /login meta. Models sourced from models.dev; default muse-spark-1.3.
Extract Google catalog processing and apply models.dev reasoning effort
metadata directly for Google, Vertex, and OpenCode models. Keep the
provider-specific Gemma 4 toggle mapping.
This change makes system prompt text and tool changes part of the transcript rather than silently rewriting its starting conditions. This lets Pi record when instructions changed or tools became available, restore that state after resuming or navigating branches, and preserve cached prompt prefixes where the upstream supports it.
Send x-session-id for OpenRouter Chat Completions and Anthropic Messages requests while preserving explicit opt-out and cache-disabled behavior.
closes#9102
Enable unsigned thinking replay and metadata-driven adaptive effort for Fireworks Messages models, with verified fallbacks for incomplete catalog metadata. Remove redundant GLM 5.2 and Kimi K3 effort aliases and add regression tests.
closes#9323
Add grok-build-0.1 to XAI_BUILTIN_EXCLUDED_MODEL_IDS so the generated xAI
catalog only lists grok-4.3, grok-4.5 and grok-4.6. Update the exclusion
test and drop the thinking-level assertion for the removed model.
Co-authored-by: Jaaneek <Jaaneek@users.noreply.github.com>
models.dev's cloudflare-ai-gateway catalog no longer lists workers-ai/*
entries, which removed every openai-completions model from the generated
gateway catalog and narrowed the inferred provider api union. Derive the
gateway's workers-ai models from the cloudflare-workers-ai catalog and pin
the provider's api generic. Also move the zai glm-5.3 cost assertion, now
that models.dev prices it in the zai API catalog.
#7638 hardcoded vLLM's thinking_token_budget. Map thinkingBudgets through
thinkingTokenBudgetField, keep the boolean as a vLLM alias, clamp so the
answer keeps 1024 tokens, and document it.
fixes#8274
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Mario Zechner <badlogicgames@gmail.com>
DEEPSEEK_V4_FLASH_THINKING_LEVEL_MAP (which enables the low effort level)
was only applied to deepseek/deepseek-v4-flash, so the same model served
through opencode and opencode-go only offered off/high/max. models.dev
reports reasoning_options effort values of low/high/max for these
providers, and the live opencode-go API accepts reasoning_effort=low
(verified against https://opencode.ai/zen/go/v1).
Apply the flash map to every deepseek-v4-flash variant on the deepseek,
opencode, and opencode-go providers. qwen-token-plan is left unchanged
since its gateway support for the low effort level is not verified.
* feat(ai): route xAI models through Responses and default to Grok 4.6
Send built-in xAI catalog models through the Responses API with
store: false and include reasoning.encrypted_content so encrypted
reasoning is requested and replayed, including future models.dev
additions. Thinking levels come from models.dev reasoning_options
(grok-4.6 exposes xhigh); models without verified options
(grok-build-0.1) keep reasoning always on and never send unsupported
"none"/"minimal" efforts. Narrow the xAI provider to the Responses API,
send pi's runtime User-Agent on xAI requests for provider-side
attribution, and make Grok 4.6 the default xAI model.
* fix(ai): drop duplicate xAI encrypted-reasoning tests
Azure already covers the shared replay path; keep the xAI-specific include-without-effort case next to the existing request tests.
---------
Co-authored-by: Jaaneek <Jaaneek@users.noreply.github.com>
