DeepSeek V3.2 API pricing
Standard tier, blended at 3:1 input to output: $0.310 per million tokens.
DeepSeek charges $0.280 per million input tokens and $0.400 per million output tokens for DeepSeek V3.2. It is sold in 4 forms, from $0.305 to $0.840 blended; the table below shows what each costs against standard. 5 of the models nearest it in price are cheaper on a blended basis.
Every tier of DeepSeek V3.2
Blended at 3:1 input to output. The last column compares each against the standard tier.
| Source | ||||||||
|---|---|---|---|---|---|---|---|---|
| Standard Standard | deepseek-v3.2 | DeepSeek | $0.280 | $0.400 | $0.028 | $0.310 | Standard | ↗ |
| Standard | deepseek-v3.2 | OpenRouter | $0.280 | $0.420 | $0.028 | $0.315 | +2% | ↗ |
| Preview | deepseek-v3.2-exp | OpenRouter | $0.270 | $0.410 | Not listed | $0.305 | −2% | ↗ |
| Alias | deepseek-v3p2 | Fireworks AI | $0.560 | $1.68 | Not listed | $0.840 | 2.7× | ↗ |
Prices last checked: . Methodology
Which tier to use
What each costs against standard in our data, and the job it is for.
Preview and beta
deepseek-v3.2-exp is a pre-release build, listed at 2% cheaper than standard. They can change behaviour or be withdrawn at short notice, so they suit trials rather than production.
Aliases
"-latest" names point at whichever release the vendor currently serves, listed at 2.7× standard. The price can change under you when the alias moves, so budget against a pinned name.
Direct or through a reseller
Through OpenRouter, deepseek-v3.2 is 2% more than buying direct. A reseller is worth it for one bill across vendors and automatic fallback; buy direct for the vendor's own batch, caching and rate-limit terms.
Against the models nearest in price
Standard tiers, blended at a 3:1 input to output ratio.
| Source | ||||||
|---|---|---|---|---|---|---|
| DeepSeek V3.2 This model | DeepSeek | $0.280 | $0.400 | 163,840 | $0.310 | ↗ |
| DeepSeek Chat | DeepSeek | $0.280 | $0.420 | 131,072 | $0.315 | ↗ |
| DeepSeek Reasoner | DeepSeek | $0.280 | $0.420 | 131,072 | $0.315 | ↗ |
| Llama 4 Maverick | OpenRouter | $0.188 | $0.652 | 524,288 | $0.304 | ↗ |
| GPT-4o mini | OpenAI | $0.150 | $0.600 | 128,000 | $0.263 | ↗ |
| Llama 3.1 70B Instruct | OpenRouter | $0.400 | $0.400 | 131,072 | $0.400 | ↗ |
| Claude Haiku 5.5 | Anthropic | $0.100 | $0.500 | 1,000,000 | $0.200 | ↗ |
| Llama 3 8B Chat Hf | Together AI | $0.200 | $0.200 | 8,192 | $0.200 | ↗ |
| GPT-4.1 nano | OpenAI | $0.100 | $0.400 | 1,047,576 | $0.175 | ↗ |
Prices last checked: . Methodology
Questions
How much does DeepSeek V3.2 cost per million tokens?
$0.280 per million input tokens and $0.400 per million output tokens, with cached input at $0.028. At a 3:1 input to output mix that blends to $0.310. Checked 28 Oct 2026.
What would DeepSeek V3.2 cost for a real workload?
A thousand requests a day at 2,000 input and 600 output tokens each comes to $24.00 a month on the standard tier.
What is the context window for DeepSeek V3.2?
163,840 tokens of input, with up to 163,840 tokens of output.
DeepSeek Chat · DeepSeek Reasoner · Llama 4 Maverick · GPT-4o mini · Llama 3.1 70B Instruct · All DeepSeek models · Cost calculator