Skip to main content
Launching the open-source gateway

DeepSeek V3.2 vs Step 3.5 Flash

SpecDeepSeek V3.2Step 3.5 Flash
Respan Index26.3 (#96)26.5 (#95)
Blended price per 1M$0.29$0.15
Context window131K256K
Input / output per 1M$0.26 / $0.38$0.1 / $0.3
Max output66K-
Inputtexttext
Released2025-12-012026-02-12

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryDeepSeek V3.2Step 3.5 Flash
Coding8.8 DeepSeek V3.210.7 Step 3.5 Flash
Agents31.3 DeepSeek V3.223.9 Step 3.5 Flash
Reasoning27.4 DeepSeek V3.234.6 Step 3.5 Flash
Knowledge & language20.1 DeepSeek V3.224.8* Step 3.5 Flash
Human preference49.6 DeepSeek V3.244.6 Step 3.5 Flash

* Estimated: no published scores in that category. How the Respan Index works

Benchmarks both report

Step 3.5 Flash leads on 3 of the 4 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkDeepSeek V3.2Step 3.5 Flash
LMArena Elo14251393
AIME 202593.1%97.3%
AIME 202694.2%96.7%
SWE-bench Verified73.1%74.4%

FAQ

Is DeepSeek V3.2 better than Step 3.5 Flash?

Step 3.5 Flash scores 26.5 on the Respan Index (#95), ahead of DeepSeek V3.2 at 26.3 (#96). Step 3.5 Flash leads on 3 of the 4 benchmarks both report.

Which is cheaper, DeepSeek V3.2 or Step 3.5 Flash?

DeepSeek V3.2 lists at $0.26 input and $0.38 output per 1M tokens; Step 3.5 Flash at $0.1 and $0.3. Step 3.5 Flash is 1.9x cheaper per token at list price.

Which has the larger context window?

DeepSeek V3.2 accepts 131K tokens and Step 3.5 Flash 256K tokens.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.