Bonus: Top up now and we'll double your first deposit — get x2 credits instantly.

Florence-2-base: Self-Hosting & Deployment Guide

Florence-2-base is an open-weight vision-language model that reads images alongside text.

What is Florence-2-base?

Florence-2-base is an open-weight vision-language model that reads images alongside text, published by Microsoft with open weights at microsoft/Florence-2-base. LLM API does not route this model today, so there is no endpoint or code sample for it here. What this page does give you is the self-hosting picture: published weight sizes, the VRAM each quantisation needs and the commands to serve it yourself. Every figure is read from the public repository and refreshed weekly. At import it had 3,052,834 downloads on Hugging Face in the previous 30 days, under the mit licence.

How to run Florence-2-base yourself

Open weights (mit), 232M parameters. Below: how much memory the weights need at each quantization, which hardware fits them, and the commands to serve it.

Florence-2-base VRAM requirements

Precision / quantizationWeightsFigureSource
BF16 (full precision)0.46 GBEstimate232M params × 2 bytes
MLX 4-bit0.16 GBFile sizemlx-community/Florence-2-base-ft-4bit

File size = actual download size on Hugging Face. Estimate = parameter count × bytes per weight. Weights only — the KV cache for long contexts needs extra memory; the fit tables below add 15% headroom as a rule of thumb.

GPU requirements: deploy Florence-2-base on your hardware

HardwareBF16 (full precision)MLX 4-bit
NVIDIA H200141 GB1×1×
NVIDIA H10080 GB1×1×
NVIDIA A100 80GB80 GB1×1×
NVIDIA L40S48 GB1×1×

GPUs needed per variant.

HardwareMLX 4-bit
Mac, 512 GBM3 UltraFits
Mac, 192 GBM2 UltraFits
Mac, 128 GBM4 Max / M3 MaxFits
Mac, 64 GBM4 Pro / MaxFits

Assumes ~75% of unified memory is usable by the GPU.

HardwareBF16 (full precision)MLX 4-bit
AWS p5.48xlarge8× H100 · 640 GBFitsFits
AWS p4de.24xlarge8× A100 80GB · 640 GBFitsFits
AWS g6e.12xlarge4× L40S · 192 GBFitsFits
Google Cloud a3-highgpu-8g8× H100 · 640 GBFitsFits
Azure ND H100 v58× H100 · 640 GBFitsFits

Whole-instance GPU memory; public instance specs.

Serve it yourself

pip install -U vllm
vllm serve microsoft/Florence-2-base
pip install 'sglang[all]'
python -m sglang.launch_server --model-path microsoft/Florence-2-base
pip install -U mlx-lm
mlx_lm.generate --model mlx-community/Florence-2-base-ft-4bit --prompt "Hello"

Commands use the official repositories above; vLLM and SGLang expose an OpenAI-compatible endpoint.

Not on LLM.API yet

We do not route this model through our API at the moment. Browse the models you can call today — most workloads have a close match already live.

Not available on LLM API yet

We do not route this model through our API at the moment, so there is no endpoint or code snippet for it yet. Browse the models you can call today — most workloads have a close match already live.

Browse available models

When to Use — When NOT to Use

Use it if...

  • You want to run Florence-2-base on your own hardware and keep every prompt inside your network
  • The weights are published openly, so you can pin one checkpoint and keep it reproducible
  • You want to fine-tune or quantise the model rather than accept a hosted configuration
  • You process documents, screenshots or charts where the image must stay in-house

Avoid if...

  • LLM API does not route this model today, so there is no endpoint here to call
  • You have no GPU capacity: the VRAM figures below are the floor, not a suggestion
  • You want zero operations — serving, scaling and upgrades are yours to run
  • You need audio or video input; this checkpoint documents text and images only

Get one key to every model

Swap your API key. Keep your code.