Falcon-180B
Parameters
180B
Context Length
2K
Modality
Text
Architecture
Dense
License
Falcon-180B TII License and Acceptable Use Policy
Release Date
23 Sept 2023
Knowledge Cutoff
Dec 2022
API Pricing (per 1M)
Self-hosted only
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
20x RTX 4090
24GB VRAM
Datacenter
6x NVIDIA A100
80GB VRAM
Apple Silicon
4x Apple M3 Max
128GB VRAM
2,048 tokens
Consumer
20x RTX 4090
24GB VRAM
Datacenter
6x NVIDIA A100
80GB VRAM
Apple Silicon
4x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
Rank
#201
| Benchmark | Score | Rank |
|---|---|---|
General Text Text Arena | 1147 | 167 |
Rankings
Overall Rank
#201
Coding Rank
-
About Falcon-180B
Falcon-180B is TII's 180B parameter flagship causal language model pretrained on 3.5T tokens for advanced reasoning, research, and coding. It utilizes Multi-Query Attention and custom 3D parallelism for high-scale enterprise deployments.
Technical Specifications
Attention
Attention Structure
Multi-Query Attention
Attention Heads
96
Key-Value Heads
1
Attention Head Dimension
-
Position Embedding
ROPE
RoPE Theta
-
Sliding Window Attention
-
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
-
Linear Attention Ratio
-
Normalization
Layer Normalization
Activation Function
GELU
Dimensions
Hidden Dimension Size
12,288
Number of Layers
60
FFN Intermediate Size (Dense)
-
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
-
About Falcon
The TII Falcon model family comprises causal decoder-only language models (7B, 40B). Their architecture, adapted from GPT-3, integrates rotary positional embeddings, Multi-Query Attention for inference efficiency, and FlashAttention for accelerated operations. Models are trained on the RefinedWeb dataset.