Falcon 40B

Falcon-40B-Instruct is a 40B-parameter causal decoder-only model built by TII based on Falcon-40B and fine-tuned on a mixture of Baize.

huggingface.co/tiiuae/falcon-40b-instruct
Falcon 40B homepage screenshot

About Falcon 40B

Falcon-40B-Instruct is a 40B-parameter causal decoder-only model built by TII based on Falcon-40B and fine-tuned on a mixture of Baize. It is made available under Apache 2.0 as a ready-to-use instruct and chat model for text generation, with example code using Transformers and PyTorch.

Pricing is not clearly stated.

Key features

  • Architecture optimized for inference with FlashAttention and multiquery attention
  • Needs 85-100GB of memory to run inference
  • Model details list English and French, with training mostly on English data
Best for: People seeking a ready-to-use Falcon-40B-based instruct model for text generation.
Watch out for
  • May not be ideal as a starting point for further fine-tuning

Falcon 40B at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Platforms
Web
Details verified
Oct 22, 2026 · Official source
Open source
Yes
Last checked
Oct 5, 2026 · Online when checked
Listed
Oct 5, 2026

Falcon 40B alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in Models & ML Research

Well-known listings from the same topic.

All 69
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing