Skip to content

Falcon 3 7B

by TII · falcon family

7B

parameters

text-generation code-generation reasoning multilingual

Falcon 3 7B is the compact variant of TII's third-generation Falcon language model family. It delivers solid general-purpose performance across text generation, coding, reasoning, and multilingual tasks in an efficient 7B parameter package. Built by the Technology Innovation Institute in Abu Dhabi, Falcon 3 7B offers competitive quality at its size class while being lightweight enough to run on consumer GPUs with 8GB or more of VRAM.

Quick Start with Ollama

ollama run 7b-q4_K_M
Resources Ollama Hugging Face Official Page
Creator TII
Parameters 7B
Architecture transformer-decoder
Context 32K tokens
Released Dec 11, 2024
License TII Falcon License 2.0
Ollama falcon3:7b

Quantization Options

Format File Size VRAM Required Quality Ollama Tag
Q4_K_M rec 4.4 GB 6.8 GB 7b-q4_K_M
Q8_0 7.5 GB 10 GB 7b-q8_0
F16 14 GB 17.5 GB 7b-fp16

Compatible Hardware

Q4_K_M requires 6.8 GB VRAM

Compatible Hardware

HardwareVRAMTypeFitEst. Speed
Mac Studio M4 Ultra 512GB512 GBmacRuns~120 tok/s
Mac Pro M2 Ultra 192GB192 GBmacRuns~118 tok/s
Mac Studio M4 Ultra 192GB192 GBmacRuns~120 tok/s
Mac Studio M4 Max 128GB128 GBmacRuns~80 tok/s
MacBook Pro M4 Max 128GB128 GBmacRuns~80 tok/s
MacBook Pro M5 Max 128GB128 GBmacRuns~80 tok/s
NVIDIA RTX PRO 6000 Blackwell96 GBgpuRuns~282 tok/s
MacBook Pro M3 Max 96GB96 GBmacRuns~59 tok/s
Mac mini M4 Pro 64GB64 GBmacRuns~40 tok/s
Mac Studio M4 Max 64GB64 GBmacRuns~80 tok/s
MacBook Pro M4 Max 64GB64 GBmacRuns~80 tok/s
MacBook Pro M5 Max 64GB64 GBmacRuns~80 tok/s
NVIDIA RTX 6000 Ada Generation48 GBgpuRuns~141 tok/s
NVIDIA RTX A600048 GBgpuRuns~113 tok/s
NVIDIA RTX PRO 5000 Blackwell48 GBgpuRuns~141 tok/s
Mac mini M4 Pro 48GB48 GBmacRuns~40 tok/s
MacBook Pro M3 Max 48GB48 GBmacRuns~59 tok/s
MacBook Pro M4 Pro 48GB48 GBmacRuns~40 tok/s
MacBook Pro M4 Max 48GB48 GBmacRuns~80 tok/s
MacBook Pro M5 Max 48GB48 GBmacRuns~60 tok/s
MacBook Pro M5 Pro 48GB48 GBmacRuns~40 tok/s
Mac Studio M4 Max 36GB36 GBmacRuns~80 tok/s
MacBook Pro M3 Pro 36GB36 GBmacRuns~22 tok/s
MacBook Pro M5 Max 36GB36 GBmacRuns~60 tok/s
NVIDIA RTX 5000 Ada Generation32 GBgpuRuns~106 tok/s
NVIDIA GeForce RTX 509032 GBgpuRuns~264 tok/s
iMac M4 32GB32 GBmacRuns~18 tok/s
Mac mini M4 32GB32 GBmacRuns~18 tok/s
MacBook Air M4 32GB32 GBmacRuns~18 tok/s
MacBook Air M5 32GB32 GBmacRuns~18 tok/s
MacBook Pro M5 32GB32 GBmacRuns~18 tok/s
AMD Radeon RX 7900 XTX24 GBgpuRuns~141 tok/s
NVIDIA GeForce RTX 3090 Ti24 GBgpuRuns~148 tok/s
NVIDIA GeForce RTX 309024 GBgpuRuns~138 tok/s
NVIDIA GeForce RTX 409024 GBgpuRuns~148 tok/s
NVIDIA RTX A500024 GBgpuRuns~113 tok/s
iMac M3 24GB24 GBmacRuns~15 tok/s
Mac mini M2 24GB24 GBmacRuns~15 tok/s
Mac mini M4 Pro 24GB24 GBmacRuns~40 tok/s
MacBook Air M2 24GB24 GBmacRuns~15 tok/s
MacBook Air M4 24GB24 GBmacRuns~18 tok/s
MacBook Air M5 24GB24 GBmacRuns~18 tok/s
MacBook Pro M4 Pro 24GB24 GBmacRuns~40 tok/s
MacBook Pro M5 24GB24 GBmacRuns~18 tok/s
MacBook Pro M5 Pro 24GB24 GBmacRuns~40 tok/s
AMD Radeon RX 7900 XT20 GBgpuRuns~118 tok/s
NVIDIA RTX 4000 Ada Generation20 GBgpuRuns~53 tok/s
MacBook Pro M3 Pro 18GB18 GBmacRuns~22 tok/s
AMD Radeon RX 6800 XT16 GBgpuRuns~75 tok/s
AMD Radeon RX 6900 XT16 GBgpuRuns~75 tok/s
AMD Radeon RX 7800 XT16 GBgpuRuns~92 tok/s
AMD Radeon RX 9060 XT 16GB16 GBgpuRuns~79 tok/s
AMD Radeon RX 9070 XT16 GBgpuRuns~96 tok/s
AMD Radeon RX 907016 GBgpuRuns~79 tok/s
Intel Arc A77016 GBgpuRuns~82 tok/s
NVIDIA GeForce RTX 4060 Ti 16GB16 GBgpuRuns~42 tok/s
NVIDIA GeForce RTX 4070 Ti Super16 GBgpuRuns~99 tok/s
NVIDIA GeForce RTX 4080 Super16 GBgpuRuns~108 tok/s
NVIDIA GeForce RTX 408016 GBgpuRuns~105 tok/s
NVIDIA GeForce RTX 5060 Ti 16GB16 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 5070 Ti16 GBgpuRuns~132 tok/s
NVIDIA GeForce RTX 508016 GBgpuRuns~141 tok/s
NVIDIA RTX A400016 GBgpuRuns~66 tok/s
iMac M1 16GB16 GBmacRuns~10 tok/s
iMac M4 16GB16 GBmacRuns~18 tok/s
Mac mini M1 16GB16 GBmacRuns~10 tok/s
Mac mini M4 16GB16 GBmacRuns~18 tok/s
MacBook Air M2 16GB16 GBmacRuns~15 tok/s
MacBook Air M3 16GB16 GBmacRuns~15 tok/s
MacBook Air M4 16GB16 GBmacRuns~18 tok/s
MacBook Air M5 16GB16 GBmacRuns~18 tok/s
MacBook Pro M1 16GB16 GBmacRuns~10 tok/s
MacBook Pro M2 Pro 16GB16 GBmacRuns~29 tok/s
MacBook Pro M5 16GB16 GBmacRuns~18 tok/s
AMD Radeon RX 6700 XT12 GBgpuRuns~56 tok/s
AMD Radeon RX 7700 XT12 GBgpuRuns~64 tok/s
Intel Arc B58012 GBgpuRuns~67 tok/s
NVIDIA GeForce RTX 3060 12GB12 GBgpuRuns~53 tok/s
NVIDIA GeForce RTX 3080 12GB12 GBgpuRuns~134 tok/s
NVIDIA GeForce RTX 4070 Super12 GBgpuRuns~74 tok/s
NVIDIA GeForce RTX 4070 Ti12 GBgpuRuns~74 tok/s
NVIDIA GeForce RTX 407012 GBgpuRuns~74 tok/s
NVIDIA GeForce RTX 507012 GBgpuRuns~99 tok/s
NVIDIA GeForce GTX 1080 Ti11 GBgpuRuns~71 tok/s
NVIDIA GeForce RTX 2080 Ti11 GBgpuRuns~91 tok/s
Intel Arc B57010 GBgpuRuns~56 tok/s
NVIDIA GeForce RTX 3080 10GB10 GBgpuRuns~112 tok/s
AMD Radeon RX 6600 XT8 GBgpuRuns~38 tok/s
AMD Radeon RX 76008 GBgpuRuns~42 tok/s
Intel Arc A7508 GBgpuRuns~75 tok/s
AMD Radeon RX 9060 XT 8GB8 GBgpuRuns~40 tok/s
NVIDIA GeForce GTX 10708 GBgpuRuns~38 tok/s
NVIDIA GeForce RTX 2060 Super8 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 2070 Super8 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 2080 Super8 GBgpuRuns~73 tok/s
NVIDIA GeForce RTX 30508 GBgpuRuns~33 tok/s
NVIDIA GeForce RTX 3060 Ti8 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 30708 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 40608 GBgpuRuns~40 tok/s
NVIDIA GeForce RTX 4060 Ti 8GB8 GBgpuRuns~42 tok/s
NVIDIA GeForce RTX 50508 GBgpuRuns~33 tok/s
NVIDIA GeForce RTX 5060 Ti 8GB8 GBgpuRuns~66 tok/s
NVIDIA GeForce RTX 50608 GBgpuRuns~49 tok/s
MacBook Air M1 8GB8 GBmacRuns~10 tok/s
MacBook Air M2 8GB8 GBmacRuns~15 tok/s
NVIDIA GeForce GTX 1660 Super6 GBgpuCPU Offload~15 tok/s
NVIDIA GeForce RTX 20606 GBgpuCPU Offload~15 tok/s

Benchmark Scores

68.0
mmlu