Up to date
tiiuae Apache 2.0 out 4 years ago
7.22B parameters, text in, text out.
Verified 2026-10-25 Rank #187 in LLMs · score 89.7 method v1 Sources 3 Downloads 30d 189.8k · #254 of 619 tracked
Picked by rule from 20 tracked pieces: the lab's own post, the most shared, a field report, the latest.
On 13 May, 2024, the Technology Innovation Institute released Falcon 2, an open-source model series where Falcon 2 11B scores 64.28 on the Hugging Face leaderboard, narrowly trailing Google's Gemma 7B at 64.29.
The 3 picks above, plus these 17. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
No API provider serves it, so its downloads are people running it themselves: 13 quantized builds and 100 Spaces.
Top Spaces running it: pliny-the-prompter/obliteratus ♥ 621 · h2oai/h2ogpt-chatbot ♥ 295 · h2oai/h2ogpt-chatbot2 ♥ 167 · tomg-group-umd/Binoculars ♥ 76
Weights only, an estimate: leave room for the KV cache. Method
About 4.3 GB at INT4, 8.7 GB at FP8, 17 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2023-04-24), amber triangles are 8 dated articles (each links to it), grey ticks are repo commits.
349 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | FalconForCausalLM |
| Model type | falcon |
| Dtype | bfloat16 |
| Params by dtype | BF16: 7,217,189,760 |
| Size band | M |
| MoE | no |
| Languages | en |
| Papers | arXiv:2205.14135, arXiv:1911.02150, arXiv:2101.00027, arXiv:2005.14165, arXiv:2104.09864, arXiv:2306.01116 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · FalconForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · FalconForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for falcon-7b: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 17 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 8.7 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.3 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
finetune: 47quantized: 13adapter: 289
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| OpenLLM-France/Claire-7B-0.1 | finetune | — | 6.92B | M | cc-by-nc-sa-4.0 | OpenLLM-France | 47 | 9k |
| maddes8cht/OpenLLM-France-Claire-7B-0.1-gguf | quantized | gguf | — | — | cc-by-nc-sa-4.0 | maddes8cht | 1 | 8.4k |
| dfurman/Falcon-7B-Chat-v0.1 | adapter | — | — | — | apache-2.0 | dfurman | 43 | 2.6k |
| mradermacher/falcon-7b-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 4.4k |
| tensorblock/falcon-7b-GGUF | quantized | GGUF | — | — | apache-2.0 | tensorblock | 0 | 3.6k |
| polyverse/falcon-7b-Q5_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | polyverse | 1 | 2.1k |
| mradermacher/falcon-7b-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 2k |
| Sandiago21/falcon-7b-prompt-answering | quantized | — | — | — | other | Sandiago21 | 2 | 1.3k |
| monsterapi/falcon_7b_3epoch_norobots | adapter | — | — | — | apache-2.0 | monsterapi | 1 | 1.1k |
| heliosbrahma/falcon-7b-sharded-bf16-finetuned-mental-health-conversational | finetune | -bf16 | — | — | mit | heliosbrahma | 7 | 807 |
| BlackUnicornSec/Basileak | adapter | GGUF | 6.92B | M | apache-2.0 | BlackUnicornSec | 3 | 798 |
| DevQuasar-6/tiiuae.falcon-7b-GGUF | quantized | GGUF | — | — | — | DevQuasar-6 | 0 | 943 |
| OpenLLM-France/Claire-7B-Apache-0.1 | finetune | — | — | — | apache-2.0 | OpenLLM-France | 7 | 586 |
| bnsapa/faq-llm | adapter | — | — | — | — | bnsapa | 0 | 917 |
| polyverse/falcon-7b-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | polyverse | 1 | 567 |
| kaetemi/falcon-7b-Q4_0-GGUF | quantized | Q4_0 | — | — | apache-2.0 | kaetemi | 0 | 558 |
| Menouar/falcon7b-linear-equations | adapter | — | — | — | apache-2.0 | Menouar | 1 | 497 |
| OpenLLM-France/Claire-7B-EN-0.1 | finetune | — | 6.92B | M | cc-by-nc-sa-4.0 | OpenLLM-France | 1 | 496 |
| BramVanroy/falcon-7b-ft-mc4_nl_cleaned_tiny | finetune | — | 6.92B | M | apache-2.0 | BramVanroy | 0 | 545 |
| hipnologo/falcon-7b-qlora-finetune-chatbot | adapter | — | — | — | apache-2.0 | hipnologo | 1 | 455 |
| monsterapi/falcon_7b_DolphinCoder | adapter | — | — | — | apache-2.0 | monsterapi | 0 | 399 |
| monsterapi/falcon_7b_norobots | adapter | — | — | — | apache-2.0 | monsterapi | 2 | 298 |
| itsmomo300/falcon-7b-sharded-bf16-finetuned-mental-health-conversational | adapter | -bf16 | — | — | apache-2.0 | itsmomo300 | 0 | 379 |
| UrFavB0i/Fine-tuned-Falcon7B-skincare-chatbot | finetune | — | 6.92B | M | apache-2.0 | UrFavB0i | 1 | 316 |
| Mastane/falcon-7b-sft-lora | finetune | — | — | — | apache-2.0 | Mastane | 1 | 305 |
| Mastane/falcon-7b-dpo-lora | finetune | — | — | — | apache-2.0 | Mastane | 1 | 305 |
| dawveed/AWS-Sage | adapter | — | — | — | apache-2.0 | dawveed | 3 | 185 |
| pmysl/falcon-7b-alpaca-pl-qlora | adapter | — | — | — | apache-2.0 | pmysl | 1 | 260 |
| ck46/falcon-7b-math-geometry-lora | adapter | — | — | — | — | ck46 | 1 | 259 |
| microsoft/falcon-7B-onnx | quantized | — | — | — | apache-2.0 | microsoft | 6 | 0 |
| sasuface/mechanic-falcon-7b | adapter | — | — | — | — | sasuface | 3 | 150 |
| EnDevSols/falcon-7b-chat-icd | adapter | — | — | — | — | EnDevSols | 1 | 224 |
| monsterapi/falcon-7b-python-code-instructions-18k-alpaca | adapter | — | — | — | apache-2.0 | monsterapi | 1 | 204 |
| chathuru/cicids2018-falcon7b | finetune | — | — | — | apache-2.0 | chathuru | 0 | 237 |
| pkarypis/falcon-lima | finetune | — | 6.92B | M | apache-2.0 | pkarypis | 0 | 224 |
| kaetemi/falcon-7b-Q8_0-GGUF | quantized | Q8_0 | — | — | apache-2.0 | kaetemi | 0 | 217 |
| TopperThijs/Falcon-narration-10-15 | adapter | — | — | — | apache-2.0 | TopperThijs | 0 | 198 |
| ck46/falcon-7b-math-lora | adapter | — | — | — | — | ck46 | 1 | 144 |
| devajsanghvi15/genAI-project | finetune | — | — | — | mit | devajsanghvi15 | 0 | 190 |
| ck46/falcon-7b-math-algebra-lora | adapter | — | — | — | — | ck46 | 1 | 135 |
| Date | Commit | SHA |
|---|---|---|
| 2024-10-12 | Adding `safetensors` variant of this model (#106) | ec89142 |
| 2024-09-28 | enh: add link to updated model | 545fd35 |
| 2023-09-29 | Move to in-library checkpoint (for real this time) (#84) | 898df13 |
| 2023-08-30 | Update README.md | f779652 |
| 2023-07-13 | Revert in-library PR (#66) | 3783374 |
| 2023-07-12 | Move to in-library checkpoint (#57) | 4e2d06f |
| 2023-06-09 | Update README.md (#19) | 2f5c3cd |
| 2023-06-09 | Add recommendations for inference and finetuning | 7327008 |
| 2023-06-05 | Update citation info | a0395a0 |
| 2023-06-05 | Update metadata | f6d4744 |
| 2023-05-31 | Remove TII Falcon LLM license | da8d49a |
| 2023-05-31 | Update license info to Apache 2.0 | 21be553 |
| 2023-05-31 | `license: apache-2.0` (#15) | d743f4b |
| 2023-05-31 | Register all auto classes (#12) | 2e35869 |
| 2023-05-30 | license: tii-falcon-llm | 87c4c0d |
| 2023-05-30 | Update config.json | f3185ff |
| 2023-05-30 | Update modelling_RW.py | 1ba2370 |
| 2023-05-30 | Update README.md (#1) | ccbff8f |
| 2023-05-26 | Add PyTorch 2.0 notice | 308ced2 |
| 2023-05-26 | Add link to RefinedWeb | 650217d |
| 2023-05-26 | Fix typo | c1a49e6 |
| 2023-05-26 | Update model card | 591607b |
| 2023-05-25 | Create README.md | 555b780 |
| 2023-05-25 | Create LICENSE.txt | 15f1e0e |
| 2023-05-25 | Update modelling_RW.py | 552521c |
| 2023-05-24 | Upload tokenizer | b65214c |
| 2023-05-24 | Update modelling_RW.py | 854fab6 |
| 2023-04-24 | Upload RWForCausalLM | 9562025 |
| 2023-04-24 | Upload RWForCausalLM | 80b61a6 |
| 2023-04-24 | initial commit | a76587e |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:36 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:04 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
