FastFrames-L

Parameters

7B

Context Length

-

Modality

Text

Architecture

-

KV Cache Rate

128 KB/tok

License

apple-amlr

Release Date

16 Nov 2026

Knowledge Cutoff

-

API Pricing (per 1M)

Self-hosted only

Hardware & System Requirements

Precision

Native

Context

Workload Memory: 6.76 GB

INT4 · 8K context

Base Weights: 3.26 GB

KV Cache: 1.00 GB

Runtime & Graphs: 2.50 GB

Minimum Required Hardware

1x NVIDIA L4 24GB

6.8 / 21.6 GB usable

1x RTX 3060 12GB

6.8 / 10.8 GB usable

Apple M4 24GB

6.8 / 18.0 GB usable

Determine requirements for your specific workload:

Open in VRAM Calculator

Runtime requirements

Checkpoint format

FP16

CUDA

>= 11.8

PyTorch

>= 2.1

Transformers

>= 4.40.1

ArchitecturePrecisionAttention kernel

Blackwell, Hopper

Native FP16

FlashAttention-3

Ada Lovelace, Ampere

Native FP16

FlashAttention-2

Architecture Diagram

Input TokensToken Embeddingx N layersNormPre-AttentionSelf-Attention+NormPre-FFNFeed-Forward NetworkActivation+Final NormOutput Logits

Evaluation Benchmarks

No evaluation benchmarks for FastFrames-L available.

Rankings

Overall Rank

-

Coding Rank

-

About FastFrames-L

FastFrames-L is an efficient vision-language foundation model developed by Apple designed for real-time video understanding and frame-level reasoning. It balances visual perception speed with multimodal dialogue accuracy on consumer hardware.

Specifications

Dimensions

Base Precision

FP16

About FastFrames

The FastFrames model family developed by Apple.


Other FastFrames Models
  • No related models available