Image-Text-to-Text
Transformers
Safetensors
inkling2_loop
conversational
Mixture of Experts
8-bit precision
modelopt
Instructions to use thinkingmachines/Inkling-2-Mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use thinkingmachines/Inkling-2-Mini with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="thinkingmachines/Inkling-2-Mini") 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("thinkingmachines/Inkling-2-Mini") model = AutoModelForMultimodalLM.from_pretrained("thinkingmachines/Inkling-2-Mini", 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]:])) - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use thinkingmachines/Inkling-2-Mini with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "thinkingmachines/Inkling-2-Mini" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "thinkingmachines/Inkling-2-Mini", "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/thinkingmachines/Inkling-2-Mini
- SGLang
How to use thinkingmachines/Inkling-2-Mini 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 "thinkingmachines/Inkling-2-Mini" \ --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": "thinkingmachines/Inkling-2-Mini", "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 "thinkingmachines/Inkling-2-Mini" \ --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": "thinkingmachines/Inkling-2-Mini", "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 thinkingmachines/Inkling-2-Mini with Docker Model Runner:
docker model run hf.co/thinkingmachines/Inkling-2-Mini
chef's kiss
#25 opened about 1 hour ago
by
szilard995
Multi-GPU llama.cpp benchmarks β 2Γ RTX 5060 Ti 16GB: pp/tg ladder, max context up to 393k, parallel tests up to 64k (UD-Q4_K_XL)
#24 opened about 3 hours ago
by
rafaelsobek
Pls release cheap mech interp tools to monitor this model
#23 opened about 3 hours ago
by
john1248
Basically just Qwen if they knew how to avoid MoE collapse
2
#22 opened about 6 hours ago
by
tabbytobias2
Tool calling: model plans an ask_user call, writes a lead-in sentence, then ends its turn without the call
#21 opened about 9 hours ago
by
janabimustafa
vLLM Inkling-2-Mini NVFP4 on a single RTX 3090 (24 GB): working port, 128k context, DSpark
1
#20 opened about 9 hours ago
by
hiutaky
Review in real world tasks?
1
#19 opened about 10 hours ago
by
DuongLeVan
CMF build for cortiq: Inkling-2-Mini on CUDA, Vulkan, Metal and CPU (one 14.7 GB file)
#18 opened about 10 hours ago
by
infosave
GLM-5.3 Flash vs Unsloth UD-Q2_K_XL: Japanese house floating in clouds three.js
π₯ 1
#17 opened about 12 hours ago
by
tiwari866
reddit user reporting 'over-writing' (as opposed to over-thinking) Bash commands and I'm experiencing the same
1
#16 opened about 13 hours ago
by
pathosethoslogos
Dinosaur problem
9
#15 opened about 13 hours ago
by
ashuezy
Surprisingly capable model for its size. A few problems.
#14 opened about 14 hours ago
by
stev236
Q8 is unable to run in last updated LM Studio with muttiple gr.cards: CUDA0 error
π 1
#13 opened about 15 hours ago
by
MrEmergentChaos
What Language
π 1
#12 opened about 16 hours ago
by
Anonim64
MLX builds for Apple Silicon (with DSpark), 16β24 GB Macs included
#11 opened about 16 hours ago
by
Paris99
This model is not available in your regionπ‘π‘π‘π‘π‘π‘π‘π‘π‘π‘π‘
π€π 2
3
#10 opened about 18 hours ago
by
inikishev
sm_120 support for dflash/speculative decoding
3
#9 opened about 18 hours ago
by
sharon8811
Ideal Sampling parameters please
2
#8 opened about 18 hours ago
by
Narutoouz
stuck repeating "actually let me act"
5
#7 opened about 23 hours ago
by
nabelekm
Installation Video and Testing - Step by Step
#6 opened 1 day ago
by
fahdmirzac
Thank you!
π€ 2
2
#5 opened 1 day ago
by
actellim
vLLM W4A16?
β 5
5
#4 opened 1 day ago
by
hiutaky
Friends, I think we can agree that this one is in the bag!
π 1
#3 opened 1 day ago
by
aarabas
nvfp4
β€οΈ 2
1
#2 opened 1 day ago
by
carinhena