OpenAI
Transcription1.55B
Transcription

Whisper Large V3

Whisper · MIT

OpenAI's best open speech-to-text model. Supports 99 languages. Near-human accuracy for English. Low VRAM requirements; runs on any GPU. Useful for builders who need voice-to-code or meeting transcription.

Parameters
1.55B
Architecture
Dense
Context
448 tokens
Released
2023-11-06
Engines
whisper.cpp, faster-whisper

Parameters

1.55B

VRAM

1.5 GB

Context

0K

Formats

3

GPUs

19

Whisper Large V3 (1.55B) requires 1.5 GB VRAM at recommended quality (Q5_K_M). At efficient quality (Q4_K_M), it fits in 1.3 GB VRAM, making it compatible with the NVIDIA RTX 4060 Laptop (40-60W). On NVIDIA Grace Blackwell Ultra GB300, expect approximately 450 tok/s at FP16. For the best experience, Starter AI Desktop ($582) is recommended.

Source: OwnRig methodology

VRAM (Recommended)

1.5 GB

Quantization

Q5_K_M

File Size

0.93 GB

Max Context

0K tokens

Primary Use

Transcription

Memory

VRAM Requirements

QualityQuantizationVRAMFile Size
fullFP163.1 GB3.1 GB
recommendedQ5_K_M1.5 GB0.93 GB
efficientQ4_K_M1.3 GB0.78 GB

Compatible GPUs

19 devices

Showing 19 of 19 entries

Hardware

Recommended Builds

Complete PC builds that can run Whisper Large V3.

FAQ

Frequently Asked Questions

How much VRAM does Whisper Large V3 need?
Whisper Large V3 requires 1.5 GB VRAM at recommended quality (Q5_K_M). At lower quality settings, it can fit in as little as 1.3 GB.
What is the best GPU for Whisper Large V3?
The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Whisper Large V3, achieving 450 tok/s at FP16 with an excellent rating.
What quantization should I use for Whisper Large V3?
For the best quality, use Q5_K_M (1.5 GB VRAM). If your GPU has limited VRAM, Q4_K_M (1.3 GB) is the most efficient option with acceptable quality.

Related Guides

All models

Data confidence: verified. Source

VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Whisper is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.