Run Whisper large-v3 on a VPS.
September 2026 · 207 of 280 plans fit · re-ranked daily
Speech-to-text via whisper.cpp; ~3 GB in F16, CPU-friendly for batch transcription. On a CPU-only VPS it needs about 4.6 GB of RAM: 3.1 GB of weights in the model's native format and 1.5 GB for the OS and runtime. 207 of the 280 plans in our index fit; the cheapest comfortable pick is Contabo's Cloud VPS 4 · 4 vCPU · 8 GB at $6.39/mo.
RAM needed · CPU inference
4.6 GB
- weights · native format
- 3.1 GB
- OS + runtime headroom
- 1.5 GB
Model card
- Size
- 1.54B parameters
- Context
- — tokens
- Kind
- speech
- Released
- 2023-11
- License
- Apache-2.0
Facts fetched from Hugging Face on 3 Sept 2026: sizes from bits-per-weight · 4,972,119 downloads.
Best VPS plans for Whisper large-v3
ranked by price among plans that fit
- 1runs comfortably$6.39/mo
- 2runs comfortably$8.72/mo
- 3runs comfortably$9.87/mo
- 4runs comfortably$10/mo
- 5runs comfortably$10/mo
- 6runs comfortably$10.12/mo
Frequently asked
- How much RAM does Whisper large-v3 need?
- About 4.6 GB for CPU inference at Q4_K_M: 3.1 GB of weights, 0 MB of KV cache at 8,192 tokens of context, and 1.5 GB of headroom. Longer contexts need more KV cache (0 KB per token for this model).
- What is the cheapest VPS that can run Whisper large-v3?
- Contabo Cloud VPS 4 · 4 vCPU · 8 GB (8 GB RAM, 4 vCPU) at $6.39/mo excl. VAT runs it comfortably as of 5 Sept 2026.
- How fast will Whisper large-v3 run on a VPS without a GPU?
- Speech models are small and CPU-friendly; throughput scales with vCPUs rather than RAM.