Skip to content
Updated minutes ago
TII

Falcon 3 7B

TII UAE · dense · 7.5B parameters · 32,768 context

Quality
50.0

Falcon 3 7B is a 7.5B parameter DENSE model from TII UAE, featuring a 32,768 token context window. With 32 transformer layers and a hidden dimension of 4,096, it delivers efficient Grouped Query Attention (GQA) for optimized inference throughput. Based on InferenceBench analysis, the optimal deployment configuration is the A10G at BF16 precision, achieving approximately 216.0 tokens/second at $0.50/million tokens.

Architecture Details

TypeDENSE
Total Parameters7.5B
Active Parameters7.5B
Layers32
Hidden Dimension4,096
Attention Heads32
KV Heads8
Head Dimension128
Vocab Size131,072

Memory Requirements

BF16 Weights

15.0 GB

FP8 Weights

7.5 GB

INT4 Weights

3.8 GB

KV-Cache per Token65536 bytes
Activation Estimate0.50 GB

Fits on (single-node)

B200 SXM BF16B100 SXM BF16GB200 NVL72 (per GPU) BF16GB300 NVL72 (per GPU) BF16H200 SXM BF16H100 SXM BF16H100 PCIe BF16H100 NVL BF16

GPU Recommendations

A10Goptimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

216.0 tok/s

Cost/Month

$285

Cost/M Tokens

$0.50

Use this config →
A30optimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

335.9 tok/s

Cost/Month

$332

Cost/M Tokens

$0.38

Use this config →
RTX 4090optimal

BF16 · 1 GPU · vllm

100/100

score

Throughput

362.9 tok/s

Cost/Month

$370

Cost/M Tokens

$0.39

Use this config →

API Pricing Comparison

No API pricing data available for this model.

Capabilities

Features

Tool Use Vision Code Math Reasoning Multilingual Structured Output

Supported Frameworks

vllmsglangtgillama-cppollama

Supported Precisions

BF16 (default)FP8INT4

Similar Models

Frequently Asked Questions

How much VRAM does Falcon 3 7B need for inference?

Falcon 3 7B requires approximately 15.0 GB of VRAM at BF16 precision, 7.5 GB at FP8, or 3.8 GB at INT4 quantization. Additional VRAM is needed for KV-cache (65536 bytes per token) and activations (~0.50 GB).

What is the best GPU for Falcon 3 7B?

The top recommended GPU for Falcon 3 7B is the A10G using BF16 precision. It achieves approximately 216.0 tokens/sec at an estimated cost of $285/month ($0.50/M tokens). Score: 100/100.

How much does Falcon 3 7B inference cost?

Falcon 3 7B inference costs vary by provider and GPU setup. Use our calculator for detailed cost estimates across all providers.