Skip to content
Get my AI grade

DeepSeek

deepseek-v3.2 benchmarks, ranks and price

Where Artificial Analysis, LiveBench, LMArena and Berkeley BFCL place deepseek-v3.2, what it costs through OpenRouter and Vercel AI Gateway, and the models either side of it on the consensus.

Updated Oct 1, 2026Full leaderboardSources and method

Read Oct 1, 2026

#30of 85

on the LLM leaderboard, with a consensus score of 58 across 2 leaderboards.

Source: Artificial Analysis, LiveBench, LMArena and Berkeley BFCL, combined by Rank.ai.

By leaderboard

Where each leaderboard places deepseek-v3.2

Artificial Analysis

Not measured by Artificial Analysis.

Intelligence Index: a composite of reasoning, knowledge, maths and coding evaluations.

LiveBench

Not measured by LiveBench.

Contamination-limited questions refreshed each release: reasoning, maths, coding, data analysis, language.

LMArena

#40of 79

Score 1425. Best run: deepseek-v3.2.

Blind head-to-head human votes, style controlled.

Berkeley BFCL

#10of 37

Score 56.73%. Best run: DeepSeek-V3.2-Exp (Prompt + Thinking).

Function calling and tool use, single turn to multi turn and web search.

Price and context

$0.93

Vercel AI Gateway: $0.62 in, $1.85 out

Nearby

The models either side of deepseek-v3.2

  1. #27GPT 5.2 (2025-12-11) High59
  2. #28Qwen3.7 Plus59
  3. #29DeepSeek V4 Pro 042358
  4. #30deepseek-v3.258
  5. #31grok-4-070956
  6. #32GPT-5.6 Luna54
  7. #33gemini-2.5-flash54

Questions about deepseek-v3.2

How good is deepseek-v3.2?
deepseek-v3.2 ranks #30 of 85 models on the Rank.ai LLM leaderboard, with a consensus score of 58 out of 100 across 2 leaderboards.
Is deepseek-v3.2 better than DeepSeek V4 Pro 0423?
Not on the consensus: DeepSeek V4 Pro 0423 scores 58 to deepseek-v3.2's 58. Compare them source by source below, since the leaderboards test different things.
Is deepseek-v3.2 better than grok-4-0709?
On the consensus, yes: 58 to 56.
How much does deepseek-v3.2 cost?
$0.62 per million input tokens and $1.85 per million output tokens on Vercel AI Gateway, the cheaper listing we read. Blended at three input tokens to one output, that is $0.93 per million.

These models recommend brands. See if they name yours.

Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.

  • Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
  • Who gets namedEvery competitor in the answers, most named first.
  • The pages AI readsThe sources behind each answer.
  • Three fixesWhat to fix first, with a brief for the first article.