## Inkling: An Open‑Weights Model

Inkling: An Open‑Weights Model

Inkling is a new language model released by Thinking Machines.

It is built on open weights, so anyone can download and use it.

The model is designed for fast inference and low memory use.

It works well on many tasks such as text generation, summarization, and code completion.

Developers can load Inkling with a few lines of code.

The following snippet shows a basic usage with the Hugging Face Transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = thinkingmachines/inkling
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

prompt = The future of AI is
inputs = tokenizer(prompt, return_tensors=pt)
output = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Inkling’s open‑weights policy encourages community research.

Researchers can fine‑tune the model for specific domains without starting from scratch.

The model’s license allows commercial use, making it a practical choice for startups and large firms alike.

For more details, see the original announcement: Inkling: Our Open‑Weights Model.