VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

DeepSeek-V4-Pro vs Qwen3 VL 235B A22B Reasoning

16 SHARED BENCHMARKS

Across 16 shared benchmarks, DeepSeek-V4-Pro scores higher on 15 and Qwen3 VL 235B A22B Reasoning on 1. The widest gap is Terminal-Bench Hard, where DeepSeek-V4-Pro scores 46.2 against 11.4. Tracked API pricing per million tokens: DeepSeek-V4-Pro $1.32 in / $3.96 out, Qwen3 VL 235B A22B Reasoning $0.40 in / $4.00 out.

DEEPSEEKVSALIBABA16 SHARED151 HEAD-TO-HEAD

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (deepseek-official, alibaba-official), otherwise the lowest tracked offer.

Compare DeepSeek-V4-Pro withEVERY TRACKED PAIRING

Anthropic12 PAIRINGS
OpenAI11 PAIRINGS
Google10 PAIRINGS
Alibaba11 PAIRINGS
DeepSeek3 PAIRINGS
Moonshot4 PAIRINGS
Meta2 PAIRINGS
Z.ai2 PAIRINGS
MiniMax1 PAIRING
NVIDIA2 PAIRINGS

Compare Qwen3 VL 235B A22B Reasoning withEVERY TRACKED PAIRING

Anthropic12 PAIRINGS
OpenAI11 PAIRINGS
Google10 PAIRINGS
Alibaba11 PAIRINGS
DeepSeek3 PAIRINGS
Moonshot4 PAIRINGS
Meta2 PAIRINGS
Z.ai2 PAIRINGS
MiniMax1 PAIRING
NVIDIA2 PAIRINGS