DeepSeek V3.1 API pricing
Standard tier, blended at 3:1 input to output: $0.875 per million tokens.
Together AI charges $0.600 per million input tokens and $1.70 per million output tokens for DeepSeek V3.1. It is sold in 5 forms, from $0.425 to $0.875 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 V3.1
Blended at 3:1 input to output. The last column compares each against the standard tier.
| Source | ||||||||
|---|---|---|---|---|---|---|---|---|
| Standard Standard | DeepSeek-V3.1 | Together AI | $0.600 | $1.70 | Not listed | $0.875 | Standard | ↗ |
| Alias | deepseek-chat-v3.1 | OpenRouter | $0.250 | $0.950 | $0.130 | $0.425 | −51% | ↗ |
| Terminus | deepseek-v3.1-terminus | OpenRouter | $0.270 | $1.00 | Not listed | $0.453 | −48% | ↗ |
| Alias | deepseek-v3p1 | Fireworks AI | $0.560 | $1.68 | Not listed | $0.840 | −4% | ↗ |
| Terminus · Alias | deepseek-v3p1-terminus | Fireworks AI | $0.560 | $1.68 | Not listed | $0.840 | −4% | ↗ |
Prices last checked: . Methodology
Which tier to use
What each costs against standard in our data, and the job it is for.
Terminus variant
The vendor also lists a terminus variant of DeepSeek-V3.1, listed at prices from 48% cheaper than standard to 4% cheaper than standard. It is the same model tuned or wired for one job, so pick one only when that is your job; otherwise standard is the reference price.
Aliases
"-latest" names point at whichever release the vendor currently serves, listed at prices from 51% cheaper than standard to 4% cheaper than 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-chat-v3.1 is 49% cheaper 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 V3.1 This model | Together AI | $0.600 | $1.70 | 131,072 | $0.875 | ↗ |
| Llama 3.1 Nemotron 70B Instruct Hf | Together AI | $0.880 | $0.880 | 32,768 | $0.880 | ↗ |
| Gemini 2.5 Flash | Google AI | $0.300 | $2.50 | 1,048,576 | $0.850 | ↗ |
| Mistral Large 3 | Mistral AI | $0.500 | $1.50 | 262,144 | $0.750 | ↗ |
| Mistral Large 4 | Mistral AI | $0.680 | $2.09 | 1,048,576 | $1.03 | ↗ |
| Mistral Large 4.0 | Mistral AI | $0.680 | $2.09 | 1,048,576 | $1.03 | ↗ |
| GPT-4.1 mini | OpenAI | $0.400 | $1.60 | 1,047,576 | $0.700 | ↗ |
| GPT-5 mini | OpenAI | $0.250 | $2.00 | 272,000 | $0.688 | ↗ |
| GPT-5.1 Codex mini | OpenRouter | $0.250 | $2.00 | 400,000 | $0.688 | ↗ |
Prices last checked: . Methodology
Questions
How much does DeepSeek V3.1 cost per million tokens?
$0.600 per million input tokens and $1.70 per million output tokens. At a 3:1 input to output mix that blends to $0.875. Checked 28 Oct 2026.
What would DeepSeek V3.1 cost for a real workload?
A thousand requests a day at 2,000 input and 600 output tokens each comes to $66.60 a month on the standard tier.
What is the context window for DeepSeek V3.1?
131,072 tokens of input, with up to 131,072 tokens of output.
Llama 3.1 Nemotron 70B Instruct Hf · Gemini 2.5 Flash · Mistral Large 3 · Mistral Large 4 · Mistral Large 4.0 · All Together AI models · Cost calculator