Image-Text-to-Text
Transformers
Safetensors
inkling2_loop
weight-edit
browser-automation
conversational
8-bit precision
modelopt
Instructions to use husky8/Inkling-2-Mini-Fast-Browser-Use with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use husky8/Inkling-2-Mini-Fast-Browser-Use with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="husky8/Inkling-2-Mini-Fast-Browser-Use") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("husky8/Inkling-2-Mini-Fast-Browser-Use") model = AutoModelForMultimodalLM.from_pretrained("husky8/Inkling-2-Mini-Fast-Browser-Use", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use husky8/Inkling-2-Mini-Fast-Browser-Use with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "husky8/Inkling-2-Mini-Fast-Browser-Use" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "husky8/Inkling-2-Mini-Fast-Browser-Use", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/husky8/Inkling-2-Mini-Fast-Browser-Use
- SGLang
How to use husky8/Inkling-2-Mini-Fast-Browser-Use with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "husky8/Inkling-2-Mini-Fast-Browser-Use" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "husky8/Inkling-2-Mini-Fast-Browser-Use", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "husky8/Inkling-2-Mini-Fast-Browser-Use" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "husky8/Inkling-2-Mini-Fast-Browser-Use", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use husky8/Inkling-2-Mini-Fast-Browser-Use with Docker Model Runner:
docker model run hf.co/husky8/Inkling-2-Mini-Fast-Browser-Use
Use the Browser Use repository name and link the verified reruns
Browse files
README.md
CHANGED
|
@@ -11,17 +11,17 @@ tags:
|
|
| 11 |
# Inkling 2 Mini Fast Browser Use
|
| 12 |
|
| 13 |
<!-- ordered-browser-release -->
|
| 14 |
-
**8.95脳 faster on WebArena鈥檚 completed-order lookup.** On an M5 Max MacBook Pro, stock took 45.89 s and the four-row edit took 5.13 s. Both returned $65.32 with the same clicks. Of the 40.76 seconds saved, 40.19 came from inference. Reversing the run order gave 8.00脳. A clean checkout measured 6.26脳 and 5.64脳. We developed the edit on this task. On RTX 5090 it passed 0/2. [Download the four-row checkpoint](https://huggingface.co/husky8/Inkling-2-Fast-
|
| 15 |
|
| 16 |
**Default download.** The main revision swaps the answer and thinking output rows. It passed 6/20 MiniWoB++ cases. Stock passed 11/20. [Task scores](https://github.com/jtc268/inkling-2-fast-computer-use#browsergym-accuracy).
|
| 17 |
|
| 18 |
-
Across 13 selected tasks that stock and an edited checkpoint both completed, mean time fell from 21.74 s to 9.64 s (2.25脳). The video uses several edits. [Watch the 68-second demo](https://jtc268.github.io/inkling-2-fast-computer-use/) 路 [Timings](https://huggingface.co/husky8/Inkling-2-Fast-
|
| 19 |
|
| 20 |
-
On RTX 5090, stock and the search-submit edit each passed 6/6 runs across three WebArena tasks. Mean time fell from 6.14 s to 3.66 s (1.67脳). [Download the search-submit checkpoint](https://huggingface.co/husky8/Inkling-2-Fast-
|
| 21 |
<!-- /ordered-browser-release -->
|
| 22 |
|
| 23 |
## Reproduce the results
|
| 24 |
|
| 25 |
-
The public harness sends visible text and numbered controls to Inkling and executes its commands in Brave. WebArena grades each run. Stock and edited weights use greedy decoding with reasoning effort set to `none`. [Setup, pinned downloads and clean-checkout results](https://github.com/jtc268/inkling-2-fast-computer-use/blob/
|
| 26 |
|
| 27 |
-
Independent modification of Thinking Machines Lab鈥檚 Inkling 2 Mini, distributed under Apache 2.0. [Original model](https://huggingface.co/thinkingmachines/Inkling-2-Mini) 路 [Model use policy](https://thinkingmachines.ai/model-acceptable-use-policy/) 路 [Modification notice](https://huggingface.co/husky8/Inkling-2-Fast-
|
|
|
|
| 11 |
# Inkling 2 Mini Fast Browser Use
|
| 12 |
|
| 13 |
<!-- ordered-browser-release -->
|
| 14 |
+
**8.95脳 faster on WebArena鈥檚 completed-order lookup.** On an M5 Max MacBook Pro, stock took 45.89 s and the four-row edit took 5.13 s. Both returned $65.32 with the same clicks. Of the 40.76 seconds saved, 40.19 came from inference. Reversing the run order gave 8.00脳. A clean checkout measured 6.26脳 and 5.64脳. We developed the edit on this task. On RTX 5090 it passed 0/2. [Download the four-row checkpoint](https://huggingface.co/husky8/Inkling-2-Fast-Browser-Use/tree/686faa4d1748f0120008ec007dc97677678e1643) 路 [Paired runs and weight audit](https://github.com/jtc268/inkling-2-fast-computer-use/blob/main/benchmark/WEBARENA-ORDER-CHECKPOINTS.md).
|
| 15 |
|
| 16 |
**Default download.** The main revision swaps the answer and thinking output rows. It passed 6/20 MiniWoB++ cases. Stock passed 11/20. [Task scores](https://github.com/jtc268/inkling-2-fast-computer-use#browsergym-accuracy).
|
| 17 |
|
| 18 |
+
Across 13 selected tasks that stock and an edited checkpoint both completed, mean time fell from 21.74 s to 9.64 s (2.25脳). The video uses several edits. [Watch the 68-second demo](https://jtc268.github.io/inkling-2-fast-computer-use/) 路 [Timings](https://huggingface.co/husky8/Inkling-2-Fast-Browser-Use/blob/main/benchmark/speedups-timings.json) 路 [All filmed outcomes](https://jtc268.github.io/inkling-2-fast-computer-use/all-runs.html).
|
| 19 |
|
| 20 |
+
On RTX 5090, stock and the search-submit edit each passed 6/6 runs across three WebArena tasks. Mean time fell from 6.14 s to 3.66 s (1.67脳). [Download the search-submit checkpoint](https://huggingface.co/husky8/Inkling-2-Fast-Browser-Use/tree/89ab0e2a2e260fdb9252702a6509b8a4490884cd).
|
| 21 |
<!-- /ordered-browser-release -->
|
| 22 |
|
| 23 |
## Reproduce the results
|
| 24 |
|
| 25 |
+
The public harness sends visible text and numbered controls to Inkling and executes its commands in Brave. WebArena grades each run. Stock and edited weights use greedy decoding with reasoning effort set to `none`. [Setup, pinned downloads and clean-checkout results](https://github.com/jtc268/inkling-2-fast-computer-use/blob/ab2a1781eb3d60b22cd32aa69e9bcf908b759fa7/benchmark/REPRODUCE-BROWSER.md).
|
| 26 |
|
| 27 |
+
Independent modification of Thinking Machines Lab鈥檚 Inkling 2 Mini, distributed under Apache 2.0. [Original model](https://huggingface.co/thinkingmachines/Inkling-2-Mini) 路 [Model use policy](https://thinkingmachines.ai/model-acceptable-use-policy/) 路 [Modification notice](https://huggingface.co/husky8/Inkling-2-Fast-Browser-Use/blob/main/NOTICE)
|