Falcon3 10B
TII's Falcon 3: competitive 10B-class under a custom license.
About
Falcon3-10B-Instruct is the Technology Innovation Institute's (Abu Dhabi) third-generation open model: 10.3B parameters, 32K context, competitive 10B-class scores (78% IFEval strict per the Open LLM Leaderboard). 10.3k HF downloads, not gated. Licensed under the Falcon LLM License — custom, non-OSI: "open weights", never "open source". The Falcon line pioneered large-scale open training (Falcon 40B/180B in 2023), but mindshare has shifted to Qwen/Mistral/Gemma, and adoption here is modest. Included as the notable non-Western, non-Chinese open-weights effort.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Open LLM Leaderboard (unverified): 78.17 IFEval strict, 38.1 MMLU-Pro — leaderboard-reported, not independently verified.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 6/10
Scored 6/10 (verified tier, metrics-v1). Evidence: HF downloads 10,269 (peer-normalized); released 668 days ago. Manual review: TII's Falcon 3 (10B, 32K, competitive IFEval); modest adoption (10.3k dl); mindshare shifted to Qwen/Mistral/Gemma. License: Falcon LLM License — custom, non-OSI; describe as 'open weights', never 'open source'.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 6.2/10
- Maintenance · recency, releases, issue load
- 5.0/10
- Openness · license + docs
- 6.5/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · HF downloads 10,269 (peer-normalized)
- · released 668 days ago
- · manual review: TII's Falcon 3 (10B, 32K, competitive IFEval); modest adoption (10.3k dl); mindshare shifted to Qwen/Mistral/Gemma