Compare DeepSeek V3.2 Exp Thinking from DeepSeek and DeepSeek V4 Flash from DeepSeek on key metrics including benchmarks, price, context length, and other model features. Access both models and hundreds of others through the AIHubMix API.
DeepSeek V3.2 Exp ThinkingDeepSeek V4 FlashThe model DeepSeek-V3.2-Exp-Think is officially named deepseek-reasoner. It is an experimental version. As an intermediate step towards the next-generation architecture, V3.2-Exp introduces DeepSeek Sparse Attention (a sparse attention mechanism) based on V3.1-Terminus, exploring and validating exploratory optimizations for training and inference efficiency on long texts.
DeepSeek V4 Flash 0423 is DeepSeek's efficiency-optimized Mixture-of-Experts text model, designed for fast inference and workflows requiring tool use, function calling, and structured outputs. It features 284B total parameters with 13B activated parameters and supports a 1M-token context window with cost-effective pricing.
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
Promotional prices show the discounted rate; see each model page for promotion windows.
Daily traffic served through AIHubMix — how demand for each model is trending.
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.
Estimate your monthly bill for the same workload on each model.
Monthly = daily × 30. Discounted rates applied where a promotion is active.
DeepSeek V3.2 Exp Thinking: $0.411/M output tokens; DeepSeek V4 Flash: $0.928/M. Use the cost calculator above to estimate your own workload.
DeepSeek V4 Flash: 1452; DeepSeek V3.2 Exp Thinking: 1438 (LMArena coding leaderboard).
DeepSeek V4 Flash: 0.7s time to first token measured on AIHubMix; see the live performance charts above for how each model behaves across the day.
DeepSeek V4 Flash at 59.0 tok/s and DeepSeek V3.2 Exp Thinking at 22.8 tok/s, measured as output throughput on AIHubMix — a separate metric from time to first token.
DeepSeek V3.2 Exp Thinking accepts text input and supports thinking, tool calling, function calling and structured outputs; DeepSeek V4 Flash accepts text input and supports tool calling, function calling and structured outputs.
Yes. AIHubMix serves every model on this page behind one OpenAI-compatible endpoint, so switching between them is a one-line change to the model field — no second account, key or SDK.
Related model match-ups readers also look at.