@qvac/sdk 0.9.2 → 0.10.1

package/NOTICE

@@ -83,46 +83,6 @@ Third-Party Model Licenses
     https://huggingface.co/personalizedrefrigerator/whisper-base-fr
   fr-tiny-ggml-model-f16
     https://huggingface.co/JaepaX/whisper-tiny-fr
-  ggml-opus-ar-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-ar-en
-  ggml-opus-de-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-de-en
-  ggml-opus-de-fr
-    https://huggingface.co/Helsinki-NLP/opus-mt-de-fr
-  ggml-opus-en-ar
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-ar
-  ggml-opus-en-es
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-es
-  ggml-opus-en-fr
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-fr
-  ggml-opus-en-it
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-it
-  ggml-opus-en-roa
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-roa
-  ggml-opus-en-ru
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-ru
-  ggml-opus-en-zh
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-zh
-  ggml-opus-es-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-es-en
-  ggml-opus-es-fr
-    https://huggingface.co/Helsinki-NLP/opus-mt-es-fr
-  ggml-opus-fr-de
-    https://huggingface.co/Helsinki-NLP/opus-mt-fr-de
-  ggml-opus-fr-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-fr-en
-  ggml-opus-fr-es
-    https://huggingface.co/Helsinki-NLP/opus-mt-fr-es
-  ggml-opus-it-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-it-en
-  ggml-opus-ja-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-ja-en
-  ggml-opus-mt-en-roa-f16
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-roa
-  ggml-opus-mt-roa-en-f16
-    https://huggingface.co/Helsinki-NLP/opus-mt-roa-en
-  ggml-opus-roa-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-roa-en
   gpt-oss-120b-Q4_K_M-00001-of-00002
     https://huggingface.co/unsloth/gpt-oss-120b-GGUF
   gpt-oss-20b-GGUF
@@ -216,14 +176,6 @@ Third-Party Model Licenses
     https://huggingface.co/nasedkinpv/parakeet-tdt-0.6b-v3-onnx-int8
   encoder-model
     https://huggingface.co/istupakov/parakeet-tdt-0.6b-v3-onnx
-  ggml-opus-en-de
-    https://huggingface.co/Helsinki-NLP/opus-mt-en-de
-  ggml-opus-en-pt
-    https://huggingface.co/Helsinki-NLP/opus-mt-tc-big-en-pt
-  ggml-opus-ru-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-ru-en
-  ggml-opus-zh-en
-    https://huggingface.co/Helsinki-NLP/opus-mt-zh-en
   parakeet-ctc-0.6b-ONNX
     https://huggingface.co/onnx-community/parakeet-ctc-0.6b-ONNX
   parakeet-rs
@@ -376,7 +328,7 @@ Third-Party Model Licenses
   model.enis.intgemm.alphas
     https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/enis
   model.enit.intgemm.alphas
-    https://github.com/mozilla/firefox-translations-models/tree/main/models/tiny/enit
+    https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/enit
   model.enja.intgemm.alphas
     https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/enja
   model.enkn.intgemm.alphas
@@ -420,7 +372,7 @@ Third-Party Model Licenses
   model.enzh.intgemm.alphas
     https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/enzh
   model.esen.intgemm.alphas
-    https://github.com/mozilla/firefox-translations-models/tree/main/models/tiny/esen
+    https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/esen
   model.eten.intgemm.alphas
     https://github.com/mozilla/firefox-translations-models/tree/main/models/base-memory/eten
   model.faen.intgemm.alphas
@@ -537,31 +489,22 @@ JavaScript Dependencies
 
   @hyperswarm/secret-stream@6.9.1
     https://github.com/holepunchto/hyperswarm-secret-stream
-  @qvac/decoder-audio@0.3.6
-    https://github.com/tetherto/qvac
   @qvac/decoder-audio@0.3.7
     https://github.com/tetherto/qvac
   @qvac/diagnostics@0.1.1
-  @qvac/diffusion-cpp@0.1.3
-    https://github.com/tetherto/qvac
-  @qvac/dl-base@0.2.1
-    https://github.com/tetherto/qvac
-  @qvac/dl-filesystem@0.2.1
+  @qvac/diffusion-cpp@0.3.0
     https://github.com/tetherto/qvac
-  @qvac/dl-hyperdrive@0.2.1
-    https://github.com/tetherto/qvac
-  @qvac/embed-llamacpp@0.13.4
+  @qvac/dl-base@0.1.1
+  @qvac/dl-hyperdrive@0.1.1
+  @qvac/embed-llamacpp@0.14.0
     https://github.com/tetherto/qvac
   @qvac/error@0.1.1
   @qvac/infer-base@0.1.1
-  @qvac/infer-base@0.2.2
-  @qvac/infer-base@0.3.1
-    https://github.com/tetherto/qvac
-  @qvac/infer-base@0.4.0
+  @qvac/infer-base@0.4.1
     https://github.com/tetherto/qvac
   @qvac/langdetect-text@0.1.2
     https://github.com/tetherto/qvac
-  @qvac/llm-llamacpp@0.14.4
+  @qvac/llm-llamacpp@0.17.3
     https://github.com/tetherto/qvac
   @qvac/logging@0.1.0
   @qvac/ocr-onnx@0.4.2
@@ -570,17 +513,17 @@ JavaScript Dependencies
     https://github.com/tetherto/qvac
   @qvac/rag@0.4.4
     https://github.com/tetherto/qvac
-  @qvac/registry-client@0.4.0
+  @qvac/registry-client@0.4.1
     https://github.com/tetherto/qvac
   @qvac/registry-schema@0.1.2
   @qvac/response@0.1.2
-  @qvac/transcription-parakeet@0.3.1
+  @qvac/transcription-parakeet@0.3.2
+    https://github.com/tetherto/qvac
+  @qvac/transcription-whispercpp@0.6.7
     https://github.com/tetherto/qvac
-  @qvac/transcription-whispercpp@0.6.1
+  @qvac/translation-nmtcpp@2.1.0
     https://github.com/tetherto/qvac
-  @qvac/translation-nmtcpp@0.6.10
-    https://github.com/tetherto/qvac-lib-infer-nmtcpp
-  @qvac/tts-onnx@0.8.2
+  @qvac/tts-onnx@0.8.5
     https://github.com/tetherto/qvac
   adaptive-timeout@1.0.1
     https://github.com/holepunchto/adaptive-timeout
@@ -612,29 +555,31 @@ JavaScript Dependencies
     https://github.com/holepunchto/bare-events
   bare-events@2.8.2
     https://github.com/holepunchto/bare-events
-  bare-fetch@2.8.1
+  bare-fetch@2.9.0
     https://github.com/holepunchto/bare-fetch
   bare-ffmpeg@1.2.2
     https://github.com/holepunchto/bare-ffmpeg
   bare-form-data@1.2.1
     https://github.com/holepunchto/bare-form-data
-  bare-fs@4.6.0
+  bare-fs@4.7.1
     https://github.com/holepunchto/bare-fs
   bare-hrtime@2.1.1
     https://github.com/holepunchto/bare-hrtime
   bare-http-parser@1.1.3
     https://github.com/holepunchto/bare-http-parser
-  bare-http1@4.5.5
+  bare-http1@4.5.6
     https://github.com/holepunchto/bare-http1
   bare-https@2.1.3
     https://github.com/holepunchto/bare-https
   bare-inspect@3.1.4
     https://github.com/holepunchto/bare-inspect
-  bare-lief@0.1.6
+  bare-lief@0.2.3
     https://github.com/holepunchto/bare-lief
-  bare-link@2.1.10
+  bare-link@3.2.1
     https://github.com/holepunchto/bare-link
-  bare-module@6.1.3
+  bare-mime@1.0.0
+    https://github.com/holepunchto/bare-mime
+  bare-module@6.2.0
     https://github.com/holepunchto/bare-module
   bare-module-lexer@1.4.7
     https://github.com/holepunchto/bare-module-lexer
@@ -648,7 +593,7 @@ JavaScript Dependencies
     https://github.com/holepunchto/bare-node
   bare-node-worker-threads@1.0.0
     https://github.com/holepunchto/bare-node
-  bare-os@3.8.7
+  bare-os@3.9.0
     https://github.com/holepunchto/bare-os
   bare-path@3.0.0
     https://github.com/holepunchto/bare-path
@@ -658,39 +603,39 @@ JavaScript Dependencies
     https://github.com/holepunchto/bare-process
   bare-rpc@1.2.0
     https://github.com/holepunchto/bare-rpc
-  bare-runtime@1.28.1
+  bare-runtime@1.28.4
     https://github.com/holepunchto/bare-runtime
-  bare-runtime-darwin-arm64@1.28.1
+  bare-runtime-darwin-arm64@1.28.4
     https://github.com/holepunchto/bare-runtime
-  bare-semver@1.0.2
+  bare-semver@1.0.3
     https://github.com/holepunchto/bare-semver
   bare-signals@4.2.0
     https://github.com/holepunchto/bare-signals
   bare-stdio@1.0.2
     https://github.com/holepunchto/bare-stdio
-  bare-stream@2.12.0
+  bare-stream@2.13.1
     https://github.com/holepunchto/bare-stream
-  bare-structured-clone@1.5.3
+  bare-structured-clone@1.5.4
     https://github.com/holepunchto/bare-structured-clone
   bare-subprocess@5.2.3
     https://github.com/holepunchto/bare-subprocess
-  bare-tcp@2.2.7
+  bare-tcp@2.2.12
     https://github.com/holepunchto/bare-tcp
   bare-thread@1.2.0
     https://github.com/holepunchto/bare-thread
-  bare-tls@2.2.1
+  bare-tls@2.2.3
     https://github.com/holepunchto/bare-tls
   bare-tty@5.1.0
     https://github.com/holepunchto/bare-tty
   bare-type@1.1.0
     https://github.com/holepunchto/bare-type
-  bare-url@2.4.0
+  bare-url@2.4.2
     https://github.com/holepunchto/bare-url
   bare-worker@4.1.6
     https://github.com/holepunchto/bare-worker
-  bare-zlib@1.3.1
+  bare-zlib@1.3.3
     https://github.com/holepunchto/bare-zlib
-  blind-relay@1.4.0
+  blind-relay@1.5.0
     https://github.com/holepunchto/blind-relay
   compact-encoding@2.19.2
     https://github.com/holepunchto/compact-encoding
@@ -700,7 +645,7 @@ JavaScript Dependencies
     https://github.com/holepunchto/events-universal
   fd-lock@2.1.1
     https://github.com/holepunchto/fd-lock
-  fs-native-extensions@1.4.5
+  fs-native-extensions@1.5.0
     https://github.com/holepunchto/fs-native-extensions
   hyperblobs@2.11.1
     https://github.com/holepunchto/hyperblobs
@@ -714,6 +659,8 @@ JavaScript Dependencies
     https://github.com/holepunchto/hypercore-storage
   hyperdb@4.22.3
     https://github.com/holepunchto/hyperdb
+  hyperdht-address@1.0.1
+    https://github.com/holepunchto/hyperdht-address
   hyperdht-stats@1.10.0
     https://github.com/holepunchto/hyperdht-stats
   hyperdispatch@1.5.1
@@ -726,7 +673,7 @@ JavaScript Dependencies
     https://github.com/holepunchto/hyperswarm-stats
   index-encoder@3.5.0
     https://github.com/holepunchto/index-encoder
-  mirror-drive@1.14.1
+  mirror-drive@1.14.2
     https://github.com/holepunchto/mirror-drive
   noise-handshake@4.2.0
     https://github.com/holepunchto/noise-handshake
@@ -798,8 +745,8 @@ JavaScript Dependencies
     https://github.com/holepunchto/corestore
   debounceify@1.1.0
     https://github.com/mafintosh/debounceify
-  dht-rpc@6.26.3
-    https://github.com/mafintosh/dht-rpc
+  dht-rpc@6.26.4
+    https://github.com/holepunchto/dht-rpc
   events@3.3.0
     https://github.com/Gozala/events
   fast-fifo@1.3.2
@@ -818,7 +765,7 @@ JavaScript Dependencies
     https://github.com/holepunchto/hypercore
   hypercore-crypto@3.6.1
     https://github.com/mafintosh/hypercore-crypto
-  hyperdht@6.29.6
+  hyperdht@6.30.0
     https://github.com/holepunchto/hyperdht
   hyperswarm@4.17.0
     https://github.com/holepunchto/hyperswarm
@@ -836,8 +783,8 @@ JavaScript Dependencies
     https://github.com/mafintosh/nat-sampler
   protocol-buffers-encodings@1.2.0
     https://github.com/mafintosh/protocol-buffers-encodings
-  protomux@3.10.1
-    https://github.com/mafintosh/protomux
+  protomux@3.10.3
+    https://github.com/holepunchto/protomux
   queue-tick@1.0.1
     https://github.com/mafintosh/queue-tick
   random-array-iterator@1.0.0

package/dist/client/api/cancel.d.ts

@@ -28,7 +28,7 @@ import { type CancelParams } from "../../schemas/index";
  * await cancel({
  *   operation: "downloadAsset",
  *   downloadKey: "download-key",
- *   delegate: { topic: "topicHex", providerPublicKey: "peerHex" },
+ *   delegate: { providerPublicKey: "peerHex" },
  * });
  *
  * @example

package/dist/client/api/cancel.js

@@ -30,7 +30,7 @@ import { InvalidResponseError, CancelFailedError } from "../../utils/errors-clie
  * await cancel({
  *   operation: "downloadAsset",
  *   downloadKey: "download-key",
- *   delegate: { topic: "topicHex", providerPublicKey: "peerHex" },
+ *   delegate: { providerPublicKey: "peerHex" },
  * });
  *
  * @example

package/dist/client/api/completion-stream.d.ts

@@ -1,19 +1,50 @@
-import { type CompletionClientParams, type CompletionStats, type McpClientInput, type Tool, type ToolCallEvent, type ToolCallWithCall, type RPCOptions } from "../../schemas/index";
+import { type CompletionClientParams, type CompletionRun, type McpClientInput, type Tool, type RPCOptions } from "../../schemas/index";
 import { type ToolInput } from "../../utils/tool-helpers";
 type CompletionParams = Omit<CompletionClientParams, "tools"> & {
     tools?: Tool[] | ToolInput[];
     mcp?: McpClientInput[];
     rpcOptions?: RPCOptions;
+    captureThinking?: boolean;
+    emitRawDeltas?: boolean;
 };
 /**
  * Generates completion from a language model based on conversation history.
  *
+ * Returns a `CompletionRun` whose canonical surfaces are:
+ *
+ *  - `events`  — `AsyncIterable<CompletionEvent>` of ordered, typed events.
+ *  - `final`   — `Promise<CompletionFinal>` with aggregated results once the
+ *                stream ends (content, thinking, tool calls, stats, raw text).
+ *
+ * Legacy convenience fields (`tokenStream`, `text`, `toolCallStream`,
+ * `toolCalls`, `stats`) are still available but deprecated — they derive
+ * from `events` / `final` internally.
+ *
  * @param params - The completion parameters
  * @param params.modelId - The identifier of the model to use for completion
  * @param params.history - Array of conversation messages with role, content, and optional attachments
  * @param params.stream - Whether to stream tokens (true) or return complete response (false). Defaults to true
  * @param params.tools - Optional array of tools (can be simple ToolInput with Zod schemas or full Tool objects)
  * @param params.mcp - Optional array of MCP client inputs for tool integration
+ * @param params.captureThinking - Best-effort parsing of `<think>` blocks into `thinkingDelta` events; `final.raw.fullText` always preserves the original output
+ * @param params.emitRawDeltas - When true, every raw model token is also emitted as a `rawDelta` event
+ * @param params.toolDialect - Override the SDK's name-based dialect detection. Use when your model emits a known format (`"hermes"`, `"pythonic"`, or `"json"`) the auto-router doesn't recognise. Drives both streaming frame detection and finalization parsing.
+ * Common override case: Llama 3.x tool-calling fine-tunes that emit the native pythonic header (`<|start_header_id|>tool_call<|end_header_id|>...<|eot_id|>`).
+ * @param params.responseFormat - Optional structured-output constraint applied to the model's output:
+ *   - `{ type: "text" }` — no constraint (default behavior)
+ *   - `{ type: "json_object" }` — output must be a JSON object
+ *   - `{ type: "json_schema", json_schema: { name, schema, description?, strict? } }` — output must validate against `schema`
+ *
+ *   The schema is converted to GBNF natively by llama.cpp and applied for the
+ *   duration of the request only. `json_schema.name` and `json_schema.description`
+ *   are accepted for OpenAI compatibility but only used at the API boundary —
+ *   they do not affect generation. **`json_schema.strict` is currently accepted
+ *   for compatibility but does NOT trigger OpenAI's auto-tightening semantics**
+ *   (implicit `additionalProperties: false`, all properties required). The
+ *   schema is forwarded to the addon as-is, so callers who want strict
+ *   validation must encode it explicitly in `schema`.
+ *
+ *   Cannot be combined with `tools` (tools already constrain output via their parameter schema).
  * @param params.kvCache - Optional KV cache configuration. Cache files are organized hierarchically:
  *   - Structure: `{kvCacheKey}/{modelId}/{configHash}.bin`
  *   - The configHash includes model config + system prompt to ensure cache isolation
@@ -22,17 +53,29 @@ type CompletionParams = Omit<CompletionClientParams, "tools"> & {
  *   - `false` or `undefined`: No caching
  *   - ⚡ Performance: When cache exists, only the last message is sent to the model (includes multimodal attachments)
  *   - 🗑️ Cleanup: Use `deleteCache({ kvCacheKey })` to remove cached sessions
- * @returns Object with tokenStream generator, toolCallStream generator, text promise, toolCalls promise (with call() method), and stats promise
+ *
+ *   **Auto-cache (`kvCache: true`) — assistant turn contract.** When
+ *   pushing the assistant turn back into `history` for the next call,
+ *   use `(await run.final).cacheableAssistantContent`. That's the exact
+ *   string the SDK persisted to the cache key on this turn, so re-using
+ *   it verbatim guarantees the next-turn lookup hits.
+ *   - Any post-processing of the assistant text (rewriting, summarizing,
+ *     stripping model stop tokens like `<|im_end|>`) before pushing it
+ *     back will miss the cache. Push the canonical string unchanged.
+ *   - `cacheableAssistantContent` is omitted on tool-call turns - those
+ *     can't be auto-cached today.
+ * @returns A CompletionRun — consume via `events` / `final`.
  * @example
  * ```typescript
  * import { z } from "zod";
  *
- * const result = completion({
+ * const run = completion({
  *   modelId: "llama-2",
  *   history: [
  *     { role: "user", content: "What's the weather in Tokyo?" }
  *   ],
  *   stream: true,
+ *   captureThinking: true,
  *   tools: [{
  *     name: "get_weather",
  *     description: "Get current weather",
@@ -45,10 +88,12 @@ type CompletionParams = Omit<CompletionClientParams, "tools"> & {
  *   }]
  * });
  *
- * for await (const token of result.tokenStream) {
- *   process.stdout.write(token);
+ * for await (const event of run.events) {
+ *   if (event.type === "contentDelta") process.stdout.write(event.text);
+ *   if (event.type === "toolCall") console.log(event.call.name, event.call.arguments);
  * }
  *
+ * const result = await run.final;
  * for (const toolCall of await result.toolCalls) {
  *   if (toolCall.invoke) {
  *     const toolResult = await toolCall.invoke();
@@ -57,12 +102,6 @@ type CompletionParams = Omit<CompletionClientParams, "tools"> & {
  * }
  * ```
  */
-export declare function completion(params: CompletionParams): {
-    tokenStream: AsyncGenerator<string>;
-    toolCallStream: AsyncGenerator<ToolCallEvent>;
-    stats: Promise<CompletionStats | undefined>;
-    text: Promise<string>;
-    toolCalls: Promise<ToolCallWithCall[]>;
-};
+export declare function completion(params: CompletionParams): CompletionRun;
 export {};
 //# sourceMappingURL=completion-stream.d.ts.map