@blank-utils/llm 0.3.2 → 0.3.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +8 -12
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -21,8 +21,10 @@
|
|
|
21
21
|
- ⚛️ **React hooks** — `useChat`, `useStream`, `useCompletion` with eager background loading
|
|
22
22
|
- 🔤 **Type-safe model selection** — full autocomplete for 30+ supported models across both backends
|
|
23
23
|
- 📝 **Streaming support** — real-time token output with abort control
|
|
24
|
+
- 📄 **PDF & Image Processing** — Extract text from PDFs natively and easily pass multimodal image attachments
|
|
24
25
|
- 🔄 **Message queueing** — users can type while models download; messages are processed once ready
|
|
25
26
|
- 🧩 **Vanilla JS friendly** — works outside React with DOM helpers and a simple `createLLM()` API
|
|
27
|
+
- ⚡ **Instant Builds** — Bundled via `tsup` for lightning-fast compilation
|
|
26
28
|
- 📦 **Zero config** — auto-detects WebGPU/WASM and picks the best backend
|
|
27
29
|
|
|
28
30
|
## Installation
|
|
@@ -408,6 +410,8 @@ import { ChatInput } from "@blank-utils/llm/react";
|
|
|
408
410
|
- ⌨️ Enter to send, Shift+Enter for newline
|
|
409
411
|
- ⏹️ Stop button while generating
|
|
410
412
|
- 🎨 Dark/light theme support
|
|
413
|
+
- 📄 **Drag-and-drop or click to upload PDF files for automatic local text extraction (OCR)**
|
|
414
|
+
- 🖼️ **Paste/upload images directly into multimodal models**
|
|
411
415
|
|
|
412
416
|
---
|
|
413
417
|
|
|
@@ -508,10 +512,6 @@ You can use either the **alias** (short name) or the **full model ID** when spec
|
|
|
508
512
|
| `qwen-2.5-7b` | Qwen 2.5 7B | Large |
|
|
509
513
|
| `qwen-2.5-coder-0.5b` | Qwen 2.5 Coder 0.5B | Code-focused |
|
|
510
514
|
| `qwen-2.5-coder-1.5b` | Qwen 2.5 Coder 1.5B | Code-focused |
|
|
511
|
-
| `qwen-3-0.6b` | Qwen 3 0.6B | Latest gen |
|
|
512
|
-
| `qwen-3-1.7b` | Qwen 3 1.7B | Latest gen |
|
|
513
|
-
| `qwen-3-4b` | Qwen 3 4B | Latest gen |
|
|
514
|
-
| `qwen-3-8b` | Qwen 3 8B | Latest gen |
|
|
515
515
|
| `gemma-2-2b` | Gemma 2 2B | Google, efficient |
|
|
516
516
|
| `gemma-2-2b-1k` | Gemma 2 2B (1K ctx) | Lower memory |
|
|
517
517
|
| `gemma-2-9b` | Gemma 2 9B | Large |
|
|
@@ -532,7 +532,6 @@ You can use either the **alias** (short name) or the **full model ID** when spec
|
|
|
532
532
|
| `qwen-2.5-1.5b` | `onnx-community/Qwen2.5-1.5B-Instruct` | Good quality |
|
|
533
533
|
| `qwen-2.5-coder-0.5b` | `onnx-community/Qwen2.5-Coder-0.5B-Instruct` | Code |
|
|
534
534
|
| `qwen-2.5-coder-1.5b` | `onnx-community/Qwen2.5-Coder-1.5B-Instruct` | Code |
|
|
535
|
-
| `qwen-3-0.6b` | `onnx-community/Qwen3-0.6B-ONNX` | Latest gen |
|
|
536
535
|
| `smollm2-135m` | `HuggingFaceTB/SmolLM2-135M-Instruct` | Ultra fast |
|
|
537
536
|
| `smollm2-360m` | `HuggingFaceTB/SmolLM2-360M-Instruct` | Fast |
|
|
538
537
|
| `smollm2-1.7b` | `HuggingFaceTB/SmolLM2-1.7B-Instruct` | Good |
|
|
@@ -586,13 +585,10 @@ bun test
|
|
|
586
585
|
|
|
587
586
|
### Build Pipeline
|
|
588
587
|
|
|
589
|
-
| Script
|
|
590
|
-
|
|
|
591
|
-
| `clean`
|
|
592
|
-
| `build
|
|
593
|
-
| `postbuild` | Copies WASM + ONNX runtime assets into `dist/` and `dist/react/` |
|
|
594
|
-
| `build:types` | Generates `.d.ts` declaration files via `tsc` |
|
|
595
|
-
| `build` | Runs all of the above in sequence |
|
|
588
|
+
| Script | What it does |
|
|
589
|
+
| ------- | ---------------------------------------------------------- |
|
|
590
|
+
| `clean` | Removes `dist/` |
|
|
591
|
+
| `build` | Super-fast bundling via `tsup`, compiling ESM code & types |
|
|
596
592
|
|
|
597
593
|
### Package Exports
|
|
598
594
|
|
package/package.json
CHANGED