Image-to-Text
Transformers
ONNX
Safetensors
vision-encoder-decoder
image-text-to-text
typst
math-ocr
formula-recognition
browser
grayscale
Instructions to use dbcccc/TypLens with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dbcccc/TypLens with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="dbcccc/TypLens")# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("dbcccc/TypLens") model = AutoModelForMultimodalLM.from_pretrained("dbcccc/TypLens", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "model_name": "TypLens V1.1", | |
| "source_model_sha256": "d42cfd2da242a11ecdb1fe8727ee01e6896a116cd46f74357c4fdce90eebe788", | |
| "server_fp32": { | |
| "split": "reused IBEM development set", | |
| "independent_test": false, | |
| "reviewer": "AI assistant", | |
| "independent_second_annotator": false, | |
| "newly_reviewed_cases": 38, | |
| "inherited_identical_image_and_prediction_cases": 97, | |
| "decoding": { | |
| "strategy": "greedy", | |
| "max_new_tokens": 1023, | |
| "forced_eos": false | |
| }, | |
| "content_review": "Compare source symbols, indices, accents and structure; allow equivalent Typst syntax and nonessential spacing.", | |
| "results": { | |
| "step": 33544, | |
| "total": 135, | |
| "correct": 127, | |
| "incorrect": 8, | |
| "unresolved": 0, | |
| "compile_success": 135, | |
| "complete_generation": 135, | |
| "exact_text_matches": 76, | |
| "exact_reference_render_matches": 88, | |
| "accepted_despite_text_difference": 51, | |
| "accepted_despite_pixel_difference": 39, | |
| "exact_text_but_content_failure": 0 | |
| } | |
| }, | |
| "browser_smoke": { | |
| "browser": "Microsoft Edge 153.0.4234.48", | |
| "provider": "wasm", | |
| "threads": 4, | |
| "images_per_variant": 3, | |
| "variants": [ | |
| "fp32", | |
| "int8" | |
| ], | |
| "completed_runs": 6, | |
| "preprocessing_matches_training": true, | |
| "tokens_match_cpu_onnx": true, | |
| "tokens_match_server_fp32_on_these_images": true, | |
| "known_content_error_present": "tilde accent recognized as hat on one smoke-test image", | |
| "full_accuracy_evaluation": false | |
| }, | |
| "export_versions": { | |
| "torch": "2.8.0+cu128", | |
| "transformers": "4.52.3", | |
| "onnx": "1.18.0", | |
| "onnxruntime": "1.22.0" | |
| }, | |
| "controlled_v1_vs_v1_1_accuracy_comparison": false, | |
| "quantized_full_accuracy_evaluation": false, | |
| "source_record_sha256": { | |
| "evaluation-full-generation-v1/completion-summary.json": "d889c42d2cb3f1333d84b77e7d86793d32660a491244be48eaf39eec50270b40", | |
| "evaluation-full-generation-v1/source-review.json": "8b2a23a519889066591ab97c952493daaadf7fd4dcdee126be66d8e5215d4ab0", | |
| "browser-final-v1/export-smoke.json": "63eb0c9acec09a71930bc8da1d300c7bcde0740cfa77f7cba12f104da2dcee77", | |
| "browser-final-v1/browser-smoke.json": "c1904f5727209744b778604ec2d7aa65dcbeec3247f34bce3b36c179d285593a" | |
| } | |
| } | |