Pyyan / Compare / Phi-4 vs DeepSeek V4-Pro vs Kimi K3

Phi-4 vs DeepSeek V4-Pro vs Kimi K3

3 of 5

Open-Weight Models · verified 13 Aug 2026

×Phi-4Microsoftcurrent
×DeepSeek V4-ProDeepSeekcurrent
×Kimi K3Moonshot AIcurrent
2 slots left
SpecificationPhi-4DeepSeek V4-ProKimi K3
SummaryBest small model — fits a 12GB GPU.Best all-round open model of 2026.Largest open-weight model shipped.
LicenceMITMITOpen weights
Parameters14B2.8T
Released17 Jul 2026
CategoryOpen-Weight ModelsOpen-Weight ModelsOpen-Weight Models
OfficialMicrosoftDeepSeekMoonshot AI

Highlighted rows are where these differ.

Phi-4

  • Beats GPT-4o on mathematics
  • Phi-4-mini runs in ~3GB at Q4

Best for local use.

Full spec sheet →

DeepSeek V4-Pro

  • Tops open leaderboards on agentic coding and reasoning
  • MIT licence — no conditions
  • Pioneered aggressive cache-hit pricing at ~$0.07/M

Best for everything open.

Full spec sheet →

Kimi K3

  • Tops the Frontend Code Arena
  • 76.8% on SWE-Bench Verified

Best for coding.

Full spec sheet →