chutes-build 1.0.4 → 1.2.0

package/CHANGELOG.md

@@ -4,7 +4,37 @@ All notable changes to Chutes Build will be documented in this file. The format
 is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this
 project follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## [Unreleased]
+## [1.2.0] - 2026-08-21
+
+### Added
+
+- **Parallel media-generation calls are capped per tool name.** image_gen/image_edit cap at 8 per model step and video-generation tools cap at 4; a burst at least twice the cap is discarded once and retried with a reminder, while any other over-cap keeps the first K and rejects the tail. Configurable via [tools.media_gen] in config.toml or CHUTES_BUILD_MAX_PARALLEL_IMAGE_GEN_CALLS / CHUTES_BUILD_MAX_PARALLEL_VIDEO_GEN_CALLS.
+
+### Changed
+
+- **The runtime is synced to upstream `1.0.5` (`d71f6e0c`).** Eleven upstream crates are added and rebranded: `xai-grok-active-sessions`, `xai-grok-foreign-sessions`, `xai-grok-session-events`, `xai-grok-session-search`, `xai-grok-home`, `xai-grok-workspace-daemon`, `xai-grok-diag-server`, `xai-grok-bundle`, `xai-compaction-transcript`, and `xai-fuzzy-file-search`. `xai-file-utils::events` and the shell's `active_sessions` / `storage/search_*` modules moved into them.
+
+### Fixed
+
+- **`models` no longer fails with "unknown ACP extension method".** The model-list client and handler had drifted onto different wire namespaces after the re-base; the client now sends what the handler registers (`chutes.build/models/list`).
+- **Worktree detection on Windows.** `WorktreeDb::get` treated only forward-slash input as a path, so a backslash path never reached the DB lookup; it now accepts both separators. `get_worktree_info` also normalises the git2 commondir back to native separators before display.
+
+## [1.1.0] - 2026-08-18
+
+### Added
+
+- **Tool calls emitted as plain text are recovered.** An open-weight model served through vLLM or SGLang can emit a well-formed tool call in its own chat-template syntax while the server-side parser returns an empty `tool_calls` array, so the turn ends having done nothing and the user sees raw markup. The runtime now recognises the Hermes/Qwen `<tool_call>`, Kimi K2 and Llama inline-function shapes (plus a strict fenced-JSON form) and turns them back into real calls only when the name resolves to a registered tool and the arguments pass that tool's own parser. `CHUTES_DISABLE_TOOL_TEXT_RECOVERY=1` disables it.
+- **`/usage` reports cache effectiveness.** A `cache_read_ratio` (cached read tokens divided by input tokens, clamped to `[0, 1]`) and per-call `cache_hit_calls` / `cache_miss_calls` counters join the token totals; the TUI usage block now shows the cached percentage and the hit count.
+
+### Changed
+
+- **Models without tool calling no longer receive a `tools` array.** The live catalog's `supported_features` now gates tool specs: a model that does not advertise `tools` gets an empty tool list instead of a request its endpoint rejects. The `StructuredOutput` tool is gated the same way and falls back to validating the final answer text directly. The default stays `true` for models absent from the catalog.
+
+### Fixed
+
+- **Parallel tool calls from providers that omit the chunk index no longer fuse.** `ToolCallDelta.index` is now optional, and the stream correlates by tool-call `id` when the index is absent; previously a missing index defaulted to `0`, merging every parallel call into one accumulator and concatenating their arguments into invalid JSON.
+- **Non-streaming requests now apply the Chutes reasoning plan.** Compaction, title generation and the advisor sent the bare payload to the configured inference base, dropping the `chat_template_kwargs` thinking switches and never routing `model-router` to the router endpoint, the same asymmetry the streaming path already handled.
+- **Stream deserialization errors name the offending chute.** A failed chunk now carries a short excerpt of the SSE payload alongside the serde error, so "expected value at line 1 column 1" says which backend produced what.
 
 ## [1.0.4] - 2026-08-12
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "chutes-build",
-  "version": "1.0.4",
+  "version": "1.2.0",
   "description": "Privacy-first, open-source coding agent for the Chutes ecosystem",
   "license": "Apache-2.0",
   "author": "Michael Gasperini (Mikesoft)",
@@ -37,12 +37,12 @@
     "check": "npm test && npm run verify:release && npm pack --dry-run"
   },
   "optionalDependencies": {
-    "chutes-build-darwin-arm64": "1.0.4",
-    "chutes-build-darwin-x64": "1.0.4",
-    "chutes-build-linux-arm64-gnu": "1.0.4",
-    "chutes-build-linux-x64-gnu": "1.0.4",
-    "chutes-build-win32-arm64": "1.0.4",
-    "chutes-build-win32-x64": "1.0.4"
+    "chutes-build-darwin-arm64": "1.2.0",
+    "chutes-build-darwin-x64": "1.2.0",
+    "chutes-build-linux-arm64-gnu": "1.2.0",
+    "chutes-build-linux-x64-gnu": "1.2.0",
+    "chutes-build-win32-arm64": "1.2.0",
+    "chutes-build-win32-x64": "1.2.0"
   },
   "keywords": [
     "ai",