Skip to content

DeepSeek V3.2

DeepSeekOpen weightsNon-reasoningVerifiedReleased Dec 1, 2025Input: text
Compare with Apodex 1.1Source scores ↗OpenRouter ↗
Overall score
50.2
Rank 97 of 222 · 36.7–63.6
Context window
128K
128,000 tokens
Price per 1M tokens
$0.28 in · $0.42 out
Source: BenchLM
Blended price
$0.32
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Coding69.1
  2. 2Instruction following#8258.2
  3. 3Reasoning56.4
  4. 4Knowledge46.7
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (6)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Agentic & tool use

Gert Labs29.57#50 of 52
Claw-Eval40.2#35 of 36
VITA-Bench18.5#9 of 11

Coding

React Native Evals71.5#13 of 15
SWE-Rebench60.9#4 of 13

Multimodal

Design Arena Website1176#67 of 91

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (BenchLM)$0.28$0.42$0.028
OpenRouter deepseek/deepseek-v3.2$0.28$0.42$0.028

Historical DeepSeek V3.2 pricing for the former `deepseek-chat` alias: $0.28 input (cache miss) / $0.028 cache-hit input / $0.42 output per million tokens. DeepSeek's published July 24, 2026 deprecation date has passed; current requests should use the V4 model IDs.

Arena ratings

Human-preference ratings from LMArena. Best-rated variant shown.

Text1425#105 · 48.1K votes
WebDev (coding)(Thinking)1362#111 · 6.9K votes

Open weights

From the Hugging Face repository.

Parameters
685B
Licence
mit
Downloads (30 days)
5.4M

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Similar score