token.app › Compare › DeepSeek V4 Pro 0423 vs MiniMax M3

DeepSeek V4 Pro 0423 vs MiniMax M3

DeepSeek V4 Pro 0423 leads on 1 of 2 shared benchmarks. MiniMax M3 is 2.3× cheaper per token on a blended 3:1 basis.

Head to head
Blended price uses a 3:1 input:output mix
DeepSeek V4 Pro 0423MiniMax M3
ProviderDeepSeekMiniMax
Input $/1M$0.955$0.300
Output $/1M$1.91$1.20
Blended $/1M$1.19$0.525
Context window1.0M1.0M
Max output384K512K
ReleasedApr 2026May 2026
LicenceOpen weightsOpen weights
GPQA Diamond 90.9% 90.9%
OTIS Mock AIME 96.7% 71.1%
Benchmark scores independently run and published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0. Only benchmarks with a score for BOTH models are compared.
Choose DeepSeek V4 Pro 0423 if…
  • Higher OTIS Mock AIME (96.7% vs 71.1%)
Choose MiniMax M3 if…
  • Costs less per token — $0.525 vs $1.19 blended
  • Newer model (released May 2026)
FAQ
Which is better, DeepSeek V4 Pro 0423 or MiniMax M3?
Across the 2 benchmarks both models have independently-run scores for, DeepSeek V4 Pro 0423 leads on 1 and MiniMax M3 leads on 0. Benchmarks are one input — price and context window are on this page too.
Is DeepSeek V4 Pro 0423 cheaper than MiniMax M3?
MiniMax M3 is cheaper: $0.525 versus $1.19 per million tokens blended at 3:1 input:output. Input alone: $0.955 vs $0.300. Output alone: $1.91 vs $1.20.
What context windows do DeepSeek V4 Pro 0423 and MiniMax M3 have?
DeepSeek V4 Pro 0423: 1,048,576 tokens. MiniMax M3: 1,048,576 tokens.
Who makes DeepSeek V4 Pro 0423 and MiniMax M3?
DeepSeek V4 Pro 0423 is made by DeepSeek. MiniMax M3 is made by MiniMax.
Related comparisons