whisper-base vs Phi-4-multimodal-instruct

Specs, VRAM requirements and download trends — updated 28 July 2026.

openai · A
Params
70 M
Context
30d
4.8 M
apache-2.0 · commercial OK
microsoft · B
Params
5.6 B
Context
131,072
30d
572 K
mit · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification whisper-base Phi-4-multimodal-instruct
Parameters 70 M 5.6 B
Architecture WhisperForConditionalGeneration Phi4MMForCausalLM
Context length 131,072
Licence apache-2.0 mit
Commercial use Allowed Allowed
Languages 99 24
Downloads 30d 4,759,784 571,625
Downloads all time 46.7 M 8.3 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2022-09-26 2025-02-24

Download trend

Daily snapshots, last 10 days (28 Jul – 6 Aug)

6.4 M539 K
whisper-base Phi-4-multimodal-instruct

VRAM side by side

On RTX 4090 · 24 GB · 8K context unless noted

whisper-base · fp16 0.8 GB / 24 GB
✅ Runs comfortably
Phi-4-multimodal-instruct · fp16 @ 128K ctx 28.0 GB / 24 GB
❌ Won’t fit

Adjacent comparisons