The Model Index
Up to four models

DeepSeek-V4-Flash (Apr 2026) vs DeepSeek-V3.2

On the index (GPT-4 = 100), DeepSeek-V4-Flash (Apr 2026) scores 142 and DeepSeek-V3.2 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to DeepSeek-V4-Flash (Apr 2026)
Presets
DeepSeek-V4-Flash (Apr 2026)
142125–159
WeightsOpen
ReleasedApr 2026
Price in / out$0.090 / $0.18
Context1.05M
Core tests taken0 of 9
DeepSeek-V3.2
141125–157
WeightsOpen
ReleasedDec 2025
Price in / out$0.26 / $0.38
Context164K
Core tests taken2 of 9

DeepSeek-V4-Flash (Apr 2026) and DeepSeek-V3.2 are too close to call (1 point apart, within the uncertainty), and costs 2.6× less.

Test by testthird-party results

DeepSeek-V4-Flash (Apr 2026)DeepSeek-V3.2
0%25%50%75%100%AGENTSDTBenchDTBenchLMCALMCA
Every result as a table
TestDeepSeek-V4-Flash (Apr 2026)DeepSeek-V3.2
Agents
DTBench86.4%85.6%
LMCA35.9%28.8%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG