DeepSeek V3.2 · DeepSeek

What does DeepSeek V3.2 cost to run in a coding agent?

DeepSeek V3.2 costs about $0.013 per average coding-agent request on our measured traffic, at OpenRouter's listed prices of $0.259 per million input and $0.80 per million output tokens. It scores 21 on the Artificial Analysis Intelligence Index (an estimate), 39th of the 48 models we track. Source: Artificial Analysis (artificialanalysis.ai)

List prices: OpenRouter listed price, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. How we compute this

Run current Claude, GPT, Gemini and open models in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

See plans

What DeepSeek V3.2 costs per coding-agent request

Cost at list price	DeepSeek V3.2
Prompt under 100K tokens50% of requests, 87% of input cached	$0.0095
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.020
Prompt over 272K tokens11% of requests, 86% of input cached	n/a
Average request	$0.013
100 requests	$1.27
Median session16 requests; a session ends after 10 idle minutes	$0.13
A month of regular usemedian of 20 users active on 5+ days	$12.93

Each logged request priced at the listed rates above for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request. 28% of requests were larger than DeepSeek V3.2's 164K context window and are left out of its figures.

DeepSeek V3.2 benchmarks, speed and latency

Artificial Analysis	Value	Rank
Intelligence Index10 evaluations, 0 to 100	21 (estimate)	39 of 48

Source: Artificial Analysis (artificialanalysis.ai). Measured variant: DeepSeek V3.2 (Reasoning), speed on DeepSeek. Fetched October 13, 2026. Rank is among the models on this site, older versions included. Artificial Analysis runs Terminal-Bench 4.0 in its own setup, so its scores differ from the tbench.ai leaderboard.

What to watch for

Its 164K-token window is smaller than 28% of our coding-agent requests, which are left out of the figures above, so long agent sessions will hit the limit. Cached input costs $0.14 per million, 54% of its input price, so agent traffic, which is mostly cache reads, gets little discount. The price is OpenRouter's listed rate on October 13, 2026, which can be a third-party host's; DeepSeek's own API may charge differently.

About DeepSeek V3.2

DeepSeek V3.2 is DeepSeek's previous-generation open-weight model, released December 1, 2025. Artificial Analysis estimates 22 on its Intelligence Index, against 39 for DeepSeek V4.1 Flash, and lists it as superseded. Its 164K context is smaller than 27.9% of the coding-agent requests we measured, and at OpenRouter's listed price it costs about twice as much per request as V4.1 Flash at DeepSeek's peak rates ($0.013 vs $0.0067).

Sources: Artificial Analysis

DeepSeek V3.2 FAQ

How much does DeepSeek V3.2 cost per coding-agent request?

$0.013 on average at OpenRouter's listed price, $1.27 per 100 requests, and $0.13 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at OpenRouter's listed rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does DeepSeek V3.2 charge more for long prompts?

No. DeepSeek V3.2 bills the same rate per token up to its 164K-token context window, which is smaller than 28% of the coding-agent requests we measured.

Is DeepSeek V3.2 good for AI agents?

Existing V3.2 setups with short prompts. On our scoring, which is calibrated to public benchmarks, it rates 4.7/10 for agentic ability and 5.2/10 for output quality. Open weights.

What is DeepSeek V3.2's context window?

DeepSeek V3.2 has a 164K-token context window. DeepSeek is the maker.

Is DeepSeek V3.2 open source?

Yes. DeepSeek V3.2 ships with open weights, so you can self-host it or run it through a hosted provider such as OpenRouter. Check the license before commercial use.

What are the downsides of DeepSeek V3.2?

The main trade-offs: 164K context, smaller than 27.9% of the coding-agent requests we measured; Superseded by DeepSeek V4.1 Flash, which scores higher for less. It's strongest for existing V3.2 setups with short prompts.