@thinletterio/vqweb 0.1.0

package/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

package/NOTICE

@@ -0,0 +1,35 @@
+thinletter embedding-quantization
+Copyright 2026 Jan Rosecký (thinletter)
+
+This product is licensed under the Apache License, Version 2.0 (see LICENSE). It includes or derives from the
+following third-party components:
+
+1. @wllama/wllama 3.6.1 (client/browser/vendor/wllama/: index.js, wllama.wasm; libllama build b10663-83d855c)
+   Copyright (c) 2024 Xuan Son NGUYEN. MIT License (client/browser/vendor/wllama/LICENCE).
+   llama.cpp / ggml inside the wasm: Copyright (c) 2023-2026 The ggml authors. MIT License.
+
+2. @huggingface/tokenizers 0.2.0 (client/vqweb/vendor/tokenizers/tokenizers.mjs)
+   Copyright 2018- The Hugging Face team. Apache License, Version 2.0 (client/vqweb/vendor/tokenizers/LICENSE).
+
+3. client/vqweb/embed.js contains a Q2_K block decoder ported operation by operation from llama.cpp
+   gguf-py/gguf/quants.py (Q2_K.dequantize_blocks). Portions derived from llama.cpp,
+   Copyright (c) 2023-2026 The ggml authors, MIT License:
+
+   Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
+   documentation files (the "Software"), to deal in the Software without restriction, including without limitation
+   the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and
+   to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above
+   copyright notice and this permission notice shall be included in all copies or substantial portions of the
+   Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT
+   LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
+   SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+   OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+   DEALINGS IN THE SOFTWARE.
+
+4. The scalar export packs GGUF files with gguf-py (llama.cpp, MIT) as an imported library; the native evaluation uses
+   llama.cpp binaries built from third_party/ (not distributed here).
+
+Methods: GPTQ (Frantar et al.), GPTVQ (van Baalen et al.), Hadamard/QuaRot-style rotations and the other borrowed ideas
+are credited in research/prior_art.md. Derived model weights published
+on Hugging Face keep the licence of their base model (microsoft/harrier-oss-v1-0.6b MIT, Qwen/Qwen3-Embedding-0.6B
+Apache-2.0, BAAI/bge-m3 MIT) with attribution in each model card.

package/README.md

@@ -0,0 +1,118 @@
+# @thinletterio/vqweb
+
+Vector-quantised **query encoders** in the browser on WebGPU. Load a released `.vqw` file (105–240 MiB), embed a query in
+about 100 ms on an integrated GPU, and search **your existing document index** — the index was built with the full-precision
+model and does not change. Apache-2.0.
+
+- Models: [huggingface.co/thinletter](https://huggingface.co/thinletter) — harrier-0.6b (English, 2.1 / 1.8 bits per weight,
+  105–129 MiB) and Qwen3-Embedding-0.6B (Czech-calibrated, 3.6 bits, 237 MiB). Each model card says which document encoder
+  and settings the file is a client of, and what it keeps of the full-precision retrieval quality (paired bootstrap on the
+  public test split; the browser reproduces the simulation).
+- Demo: [thinletter.io/demo](https://thinletter.io/demo). Method and numbers: the
+  [technical report](https://github.com/rosecky/embedding-quantization-public/blob/main/docs/release/technical_report.md).
+
+## Install
+
+```
+npm install @thinletterio/vqweb
+```
+
+or without a bundler, straight from a CDN:
+
+```html
+<script type="module">
+  import { loadVqwClient } from 'https://thinletter.io/vqweb/vqweb.js';
+  // or: https://cdn.jsdelivr.net/npm/@thinletterio/vqweb/dist/vqweb.js
+</script>
+```
+
+Requirements: WebGPU with the `shader-f16` feature (Chrome / Edge on desktop and Android; Safari 26 and Firefox where the
+adapter reports `shader-f16`; `probeWebGPU()` tells you). There is no CPU fallback: check `probeWebGPU()` first and use the scalar GGUF files from the same Hugging
+Face organisation with [wllama](https://github.com/ngxson/wllama) where WebGPU is missing.
+
+## Use
+
+```js
+import { loadVqwClient, probeWebGPU } from '@thinletterio/vqweb';
+
+const gpu = await probeWebGPU();
+if (!gpu.ok) throw new Error(gpu.reason);
+
+const client = await loadVqwClient(
+  'https://huggingface.co/thinletter/qwen3-embedding-0.6b-vq-clients/resolve/main/qwen3-0.6b-vq3.5d2-webfaq_q-vocfull.vqw',
+  { onProgress: ({ phase, loaded, total }) => console.log(phase, loaded, total) },
+);
+
+const q = await client.embed('Kolik stojí parkování v centru Prahy?');   // Float32Array(1024), L2-normalised
+// score = dot(q, docVector) for every document vector of your fp32 / int8 index built with Qwen/Qwen3-Embedding-0.6B
+
+console.log(client.info.model.teacher, client.info.quant.bpwBlocks, client.info.timing);
+client.dispose();
+```
+
+What `loadVqwClient` does: downloads the container (with progress; a `.chunks.json` byte-chunk manifest as on thinletter.io
+works too), keeps it in the origin's private file system (OPFS) so the next visit loads from disk, uploads the codebooks,
+indices and scales to the GPU, fetches the tokenizer named in the container header, compiles the WebGPU pipelines and runs
+one warm-up query. The container header carries the query prompt, the EOS rule, the byte-fallback token table and, for the
+`*p` files, the prompt's precomputed keys and values — `embed(text)` takes the **raw query text**.
+
+Options: `cache: 'none'` (memory only), `device` (bring your own `GPUDevice` with `shader-f16`), `powerPreference`
+(`'low-power'` default = the integrated GPU where there is a choice), `tokenizerUrl` / `tokenizerConfigUrl` (override the
+files next to the container), `warmup: false`. `client.embedMany(texts)` runs queries one after another (the runtime is
+batch 1). `client.embedWithStats(text)` adds the token count and the milliseconds. `clearCache(url)` removes a cached file.
+
+### In a Web Worker
+
+The package touches no DOM. Load the client in a worker to keep the page responsive during the ~1–3 s of upload and
+pipeline compilation:
+
+```js
+// worker.js
+import { loadVqwClient } from '@thinletterio/vqweb';
+let client;
+onmessage = async ({ data }) => {
+  if (data.load) { client = await loadVqwClient(data.load, { onProgress: (p) => postMessage({ progress: p }) }); postMessage({ ready: client.info }); }
+  if (data.embed) postMessage({ id: data.id, vec: await client.embed(data.embed) });
+};
+```
+
+### React
+
+```jsx
+const [client, setClient] = useState(null);
+useEffect(() => { let c; loadVqwClient(URL).then((x) => setClient(c = x)); return () => c?.dispose(); }, []);
+```
+
+## Which file
+
+| file | base model (the index must be built with it) | bits / weight | size | keeps of fp32 nDCG@10 |
+|---|---|---|---|---|
+| `harrier-0.6b-vq2.0p-scifact.vqw` | microsoft/harrier-oss-v1-0.6b | 2.10 + prompt K/V | 122 MiB | SciFact 98.1 % |
+| `harrier-0.6b-vq2.0-scidocs.vqw` | microsoft/harrier-oss-v1-0.6b | 2.10 | 127 MiB | SciDocs 94.7 % |
+| `harrier-0.6b-vq1.75p-scifact.vqw` | microsoft/harrier-oss-v1-0.6b | 1.83 + prompt K/V | 107 MiB | SciFact 95.8 % |
+| `qwen3-0.6b-vq3.5d2-webfaq_q-vocfull.vqw` | Qwen/Qwen3-Embedding-0.6B | 3.57 | 237 MiB | WebFAQ-cs 96.8 % |
+
+All files: [thinletter/harrier-0.6b-vq-clients](https://huggingface.co/thinletter/harrier-0.6b-vq-clients) and
+[thinletter/qwen3-embedding-0.6b-vq-clients](https://huggingface.co/thinletter/qwen3-embedding-0.6b-vq-clients). A file
+calibrated on one corpus family transfers to similar text; for your own index the repository has the recipe to verify
+(paired interval over your queries) and to compile a client calibrated on your corpus.
+
+## Numbers to expect
+
+Integrated Intel GPU (Xe-LPG), Chrome: 103 ms per query for the 2.1-bit harrier file, 112 ms for the 3.6-bit Qwen3 file
+(p50, ~30 tokens); load 2–3 s after the download; GPU memory = the file size plus ~40 MiB of activations. Discrete GPUs and
+phones are not measured yet. The browser output equals the exporter's simulation of the same file to ±0.003 nDCG@10.
+
+## Advanced
+
+The bundle exports the runtime modules unchanged (`parseVqw`, `VqwContainer`, `VqwRuntime`, `VqwTokenizer`,
+`requestDevice`, `KERNELS`), so a bench or a layer test can drive the GPU pass directly; `client.embedTokens(ids,
+{ profile: true })` returns per-kernel GPU timestamps where the adapter supports them. Source, format description and
+verification ladder: [`client/vqweb`](https://github.com/rosecky/embedding-quantization-public/tree/main/client/vqweb)
+in the repository; this package is built from it by `packages/vqweb/build.mjs`.
+
+## Licence
+
+Apache-2.0 (see LICENSE and NOTICE: bundles `@huggingface/tokenizers` 0.2.0, Apache-2.0, and a Q2_K decoder ported from
+llama.cpp's gguf-py, MIT). The model files carry the licence of their base model (MIT for harrier, Apache-2.0 for
+Qwen3-Embedding). Contact: info@thinletter.io

package/dist/vqweb.d.ts

@@ -0,0 +1,131 @@
+// Type declarations of @thinletter/vqweb (hand-written; the bundle is plain ES2022).
+/// <reference types="@webgpu/types" />
+
+export const VERSION: string;
+
+export interface ProgressEvent {
+  /** download | upload | tokenizer | pipelines | warmup | ready */
+  phase: string;
+  loaded: number;
+  total: number;
+}
+
+export interface LoadOptions {
+  /** 'opfs' (default): keep the container and the tokenizer in the origin's private file system for the next visit; 'none': memory only. */
+  cache?: 'opfs' | 'none';
+  onProgress?: (p: ProgressEvent) => void;
+  /** Override tokenizer.json (default: the header's file name next to the container, falling back to tokenizer.json). */
+  tokenizerUrl?: string;
+  tokenizerConfigUrl?: string;
+  /** An existing GPUDevice with the shader-f16 feature; it is not destroyed by dispose(). */
+  device?: GPUDevice;
+  /** Adapter choice when the client creates the device (default 'low-power' = the integrated GPU where there is a choice). */
+  powerPreference?: 'low-power' | 'high-performance';
+  /** One untimed query after the pipelines compile (default true). */
+  warmup?: boolean;
+  /** Advanced runtime options: resid 'f32' | 'f16', tmax, tile, usePrefix. */
+  runtime?: Record<string, unknown>;
+}
+
+export interface AdapterInfo {
+  vendor?: string; architecture?: string; device?: string; description?: string;
+  shader_f16?: boolean; subgroups?: boolean; timestamp_query?: boolean;
+  maxComputeWorkgroupStorageSize?: number; maxStorageBufferBindingSize?: number; maxBufferSize?: number; maxComputeInvocationsPerWorkgroup?: number;
+}
+
+export interface ClientInfo {
+  file: { url: string; name: string; bytes: number; cached: boolean; persisted: boolean };
+  model: { teacher: string; hf: string | null; arch: string; layers: number; hidden: number; prompt: string; addEos: boolean; maxLen: number; prefixTokens: number; pooling: string };
+  quant: { method: string; bits: number; k: number; dim: number; bpwBlocks: number | null; calib: string | null; blocksize: number };
+  tokenizer: { rows: number; full: number | null; file: string; fallback: boolean };
+  adapter: AdapterInfo | null;
+  gpuBytes: number; weightBytes: number; activationBytes: number;
+  timing: { download_s: number; upload_s: number; tokenizer_s: number; pipelines_s: number; warmup_s: number };
+  version: string;
+}
+
+export interface EmbedResult {
+  /** The raw (not normalised) pooled vector. */
+  embedding: Float32Array;
+  /** Query tokens after the stored prompt K/V were cut (if the container carries them). */
+  T: number;
+  /** Prefix tokens taken from the container's stored K/V (0 = plain path). */
+  PP: number;
+  /** Wall-clock milliseconds of the GPU pass including the readback. */
+  ms: number;
+  dispatches: number;
+  profile?: { by_kind_ms: Record<string, number>; gpu_total_ms: number; span_ms: number };
+}
+
+export declare class VqwClient {
+  readonly info: ClientInfo;
+  readonly device: GPUDevice;
+  /** Advanced handles to the bundled runtime modules. */
+  readonly container: VqwContainer;
+  readonly runtime: VqwRuntime;
+  readonly tokenizer: VqwTokenizer;
+  readonly disposed: boolean;
+  /** Query text (without the prompt: the container applies its own) -> token ids on the container's table. */
+  tokenize(text: string): number[];
+  /** Raw runtime call on token ids. */
+  embedTokens(ids: number[], opts?: { profile?: boolean; debug?: boolean }): Promise<EmbedResult>;
+  /** One query -> L2-normalised Float32Array(hidden). */
+  embed(text: string): Promise<Float32Array>;
+  /** Several queries, one after another (the runtime is batch 1). */
+  embedMany(texts: string[]): Promise<Float32Array[]>;
+  embedWithStats(text: string): Promise<{ embedding: Float32Array; tokens: number; ms: number; gpuMs: number }>;
+  /** Free the GPU buffers (and the device when this client created it). */
+  dispose(): void;
+}
+
+/** Load a .vqw container (plain URL or a <file>.chunks.json byte-chunk manifest) and prepare it for queries. */
+export function loadVqwClient(url: string, opts?: LoadOptions): Promise<VqwClient>;
+
+/** Can this browser run a client? Requests an adapter, keeps no device. */
+export function probeWebGPU(opts?: { powerPreference?: 'low-power' | 'high-performance' }): Promise<{
+  ok: boolean; reason: string | null; adapter: { vendor?: string; architecture?: string; device?: string; description?: string } | null; shaderF16: boolean | null;
+}>;
+
+/** Remove a cached container (by URL or file name) from this origin's private file system. */
+export function clearCache(urlOrName: string): Promise<boolean>;
+
+export function l2normalize(v: Float32Array | number[]): Float32Array;
+
+// ---- advanced: the runtime modules as bundled (client/vqweb in the repository) --------------------------------------------------
+export interface ParsedVqw {
+  header: any; dataStart: number; bytes: Uint8Array;
+  raw(name: string): Uint8Array;
+  view(name: string): Uint8Array | Uint16Array | Uint32Array | Int8Array | Int32Array | Float32Array;
+}
+export function parseVqw(data: ArrayBuffer | Uint8Array): ParsedVqw;
+export function fetchVqw(url: string, onProgress?: (loaded: number, total: number) => void): Promise<ArrayBuffer>;
+export function requestDevice(opts?: { powerPreference?: 'low-power' | 'high-performance'; label?: string }): Promise<{ adapter: GPUAdapter; device: GPUDevice; info: AdapterInfo }>;
+export declare class VqwContainer {
+  constructor(parsed: ParsedVqw, device: GPUDevice);
+  static load(url: string, device: GPUDevice, opts?: { onProgress?: (loaded: number, total: number) => void }): Promise<VqwContainer>;
+  readonly header: any; readonly model: any; readonly quant: any; readonly rotation: any; readonly tokens: any; readonly prefix: { ids: number[]; tokens: number; text: string } | null;
+  readonly gpuBytes: number;
+  linear(prefix: string): any; norm(name: string): GPUBuffer; rot(n: number): any; prefixKV(li: number): { k: GPUBuffer; v: GPUBuffer } | null;
+  release(): void; destroy(): void;
+}
+export declare class VqwRuntime {
+  constructor(container: VqwContainer, device: GPUDevice, opts?: Record<string, unknown>);
+  static create(container: VqwContainer, device: GPUDevice, opts?: Record<string, unknown>): Promise<VqwRuntime>;
+  readonly actBytes: number; readonly tmax: number;
+  embed(ids: number[], opts?: { profile?: boolean; debug?: boolean }): Promise<EmbedResult>;
+  destroy(): void;
+}
+export declare class VqwTokenizer {
+  constructor(tokenizerJSON: any, tokenizerConfig: any, header: any, keptIds: Iterable<number>);
+  encode(query: string): number[];
+  encodeFull(text: string): number[];
+  retokenize(ids: number[]): number[];
+  decodeOne(id: number): string;
+}
+export const KERNELS: Record<string, string>;
+export function loadModelBytes(url: string, opts?: { cache?: 'opfs' | 'none'; onProgress?: (loaded: number, total: number) => void }): Promise<{ buffer: ArrayBuffer; bytes: number; cached: boolean; persisted: boolean; name: string }>;
+export function loadTokenizerFiles(containerUrl: string, header: any, opts?: { tokenizerUrl?: string; tokenizerConfigUrl?: string; cache?: 'opfs' | 'none' }): Promise<{ tokenizerJSON: any; tokenizerConfig: any; fallback: boolean }>;
+export function fetchWithProgress(url: string, onProgress?: (loaded: number, total: number) => void, expectedBytes?: number | null): Promise<Uint8Array>;
+export function opfsGet(name: string, size: number | null): Promise<File | null>;
+export function opfsPut(name: string, parts: Array<Blob | ArrayBuffer | Uint8Array | string>): Promise<File | null>;
+export function opfsRemove(name: string): Promise<boolean>;