canirunthismodel

Can I run whisper-large-v3 locally?

openai/whisper-large-v3
Seq2Seq / translation1.5B params23 GB downloadWhisperForConditionalGeneration

whisper-large-v3 is a seq2seq / translation model with about 1.5B parameters. The practical minimum to run it is roughly 1.8 GB of GPU or system memory, and a GPU with 2 GB VRAM runs it comfortably. Here's exactly what it needs and how to run it.

Check YOUR exact machine

whisper-large-v3 memory & VRAM requirements

How much memory whisper-large-v3 needs at each quantization level (weights plus ~20% runtime overhead). Lower-bit quantization dramatically reduces VRAM at a small quality cost.

PrecisionMemory neededNotes
Full precision (FP16/BF16)4.7 GBFull quality
8-bit (INT8 / Q8)2.9 GBSmaller, slight quality loss
4-bit (Q4_K_M / INT4)2.0 GBBest size/quality trade-off
3-bit (Q3_K)1.8 GBSmaller, slight quality loss

Will whisper-large-v3 run on your GPU?

Verdicts for common setups — from CPU-only laptops to an RTX 4090 and data-center GPUs. Click a GPU to see everything it can run.

Your setupVerdictNeedsEst. speedBest way to run
No GPU (16 GB RAM)Workarounds4.7 GB~13 tok/sRun on CPU (RAM) with FP16
RTX 4060 (8 GB)Runs4.7 GB~55 tok/sRun on your GPU with FP16
RTX 3060 (12 GB)Runs4.7 GB~70 tok/sRun on your GPU with FP16
RTX 4070 Super (12 GB)Runs4.7 GB~93 tok/sRun on your GPU with FP16
RTX 4080 Super (16 GB)Runs4.7 GB~130 tok/sRun on your GPU with FP16
RTX 4090 (24 GB)Runs4.7 GB~160 tok/sRun on your GPU with FP16
RTX 3090 (24 GB)Runs4.7 GB~150 tok/sRun on your GPU with FP16
Apple M-series (18 GB unified)Runs4.7 GB~32 tok/sRun on your Apple Silicon (unified memory) with FP16
Apple M Max (64 GB unified)Runs4.7 GB~77 tok/sRun on your Apple Silicon (unified memory) with FP16
A100 (80 GB)Runs4.7 GB~240 tok/sRun on your GPU with FP16

How to run whisper-large-v3 locally

Recommended method: Run on your GPU with FP16 using vLLM or Transformers.

Serve with vLLM (OpenAI-compatible API)
# Fast production serving on http://localhost:8000/v1
vllm serve openai/whisper-large-v3
Python (Transformers)
from transformers import pipeline
pipe = pipeline("text-generation", model="openai/whisper-large-v3", device_map="auto")
print(pipe("Hello", max_new_tokens=50))

whisper-large-v3 — frequently asked questions

How much VRAM does whisper-large-v3 need?

whisper-large-v3 needs roughly 2.0 GB of VRAM at 4-bit (Q4) quantization and about 4.7 GB at full FP16 precision, including runtime overhead. Lower-bit quantization trades a little quality for a lot less memory.

Can I run whisper-large-v3 on CPU without a GPU?

Yes — whisper-large-v3 can run on CPU using system RAM (best with a quantized GGUF build via llama.cpp or Ollama), but generation will be noticeably slower than on a GPU.

What's the best way to run whisper-large-v3 locally?

The easiest path is usually vLLM or Transformers. Run on your GPU with FP16. This page's "How to run it" section has copy-paste commands.

What GPU do I need to run whisper-large-v3?

A GPU with at least 2 GB of VRAM runs whisper-large-v3 comfortably at 4-bit quantization, or about 5 GB for full FP16 precision. On Apple Silicon, unified memory of that size works too.

Related models

Check this model on a specific GPU