FitMyLLM
TII/Dense

TIIFalcon3-10B

Falcon3 10B — strong mid-range model from TII with good multilingual support.

chatThinkingTool Use
10.3B
Parameters
32K
Context length
7
Benchmarks
10
Quantizations
60K
HF downloads
Architecture
Dense
Released
2024-11-29
Layers
40
KV Heads
4
Head Dim
256
Family
falcon

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q3_K_M4
7.5 GB
5.6 + 1.9 KV
low
Q3_K_L4.3
7.9 GB
6.0 + 1.9 KV
moderate
IQ4_XS4.46
8.1 GB
6.2 + 1.9 KV
moderate
Q4_K_S4.67
8.4 GB
6.5 + 1.9 KV
moderate
Q4_K_M4.89
8.7 GB
6.8 + 1.9 KV
good
Q5_K_S5.57
9.5 GB
7.7 + 1.9 KV
good
Q5_K_M5.7
9.7 GB
7.8 + 1.9 KV
good
Q6_K6.56
10.8 GB
8.9 + 1.9 KV
excellent
Q8_08.5
13.3 GB
11.4 + 1.9 KV
lossless
FP1616
23.0 GB
21.1 + 1.9 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN FALCON3-10B NOW

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (7)

IFEval68.0
HumanEval62.0
BBH45.1
MMLU-PRO40.0
MATH27.5
MUSR14.2
GPQA10.6

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run falcon3:10b-instruct-q4_K_M

Downloads and runs automatically. Add --verbose for speed stats.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

Apple M1 (8GB)
8 GB VRAM • 68 GB/s
APPLE
$499
Apple M2 (8GB)
8 GB VRAM • 100 GB/s
APPLE
$599
Apple M3 (8GB)
8 GB VRAM • 100 GB/s
APPLE
$599
NVIDIA Tesla K8
8 GB VRAM • 160 GB/s
NVIDIA
NVIDIA Tesla M60
8 GB VRAM • 160 GB/s
NVIDIA

Find the best GPU for Falcon3-10B

Build Hardware for Falcon3-10B

Falcon3 10B — strong mid-range model from TII with good multilingual support.

▸ SPEC SHEET

Falcon3-10B10.3B Dense.

▸ SPECIFICATIONS
PARAMETERS
10.3B
ARCHITECTURE
Dense Transformer
CONTEXT LENGTH
32K tokens
CAPABILITIES
chat
RELEASE DATE
2024-11-29
PROVIDER
TII
FAMILY
falcon
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
Q3_K_M45.6 GB88%
Q3_K_L4.36.0 GB90%
IQ4_XS4.466.2 GB92%
Q4_K_S4.676.5 GB93%
Q4_K_M4.896.8 GB94%
Q5_K_S5.577.7 GB96%
Q5_K_M5.77.8 GB96%
Q6_K6.568.9 GB97%
Q8_08.511.4 GB100%
FP161621.1 GB100%
§ 01BENCHMARK SCORES
HumanEval62.0
MMLU-PRO40.0
MATH27.5
IFEval68.0
BBH45.1
GPQA10.6
MUSR14.2
§ 02RUN COMMAND

Run Falcon3-10B locally with Ollama — needs 6.8 GB VRAM at Q4_K_M:

$ollama run falcon3:10b
§ 03COMPATIBLE GPUs
30 @ Q4_K_M