Models / DeepSeek

DeepSeek V3 API pricing

$1.00 in, $4.00 out per million tokens.

deepseek-v3managed routeOpenAI SDK

Price · per million tokens, USD

Get a key
TokensDeepSeek list priceUplinkSwitchSaving
Input—$1.00—
Cached inputBilled at the input rate.
Output—$4.00—
No list price on record.
Real-model check: Steady since Oct 20

DeepSeek V3: what it is · from DeepSeek’s own model card, read 2026-10-20

The model the V3 line starts from, a mixture of experts with Multi-head Latent Attention and multi-token prediction. DeepSeek reports it holds up in needle-in-a-haystack tests across the full 128K window. It is served as deepseek-v3 at $1.00 input and $4.00 output per million tokens, on the same key as every other model in the catalog.

DeepSeek V3
DeveloperDeepSeek
Parameters671B total, 37B active per token
Context128K tokens
LicenseMIT for the code and the DeepSeek Model License for the weights, which allows commercial use
Published2024-12-25
Downloads, 30 days to 2026-10-201,377,427
Sourcedeepseek-ai/DeepSeek-V3

From deepseek-ai/DeepSeek-V3, read 2026-10-20.

curl
curl https://api.uplinkswitch.com/v1/chat/completions \
  -H "authorization: Bearer sk-uplink-live-..." \
  -H "content-type: application/json" \
  -d '{"model":"deepseek-v3","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}]}'

DeepSeek V3 next to DeepSeek’s other models · per million tokens, live

ModelHere, in / out1,000 chat turns
DeepSeek V4 Flash 0423$0.0279 / $0.1116$0.11
DeepSeek V4.1 Flash 1017$0.0437 / $0.1746$0.17
DeepSeek V4.1 Flash$0.06 / $0.18$0.21
DeepSeek V4 Flash 0731$0.084 / $0.12$0.23
DeepSeek V4 Flash Vision Exp$0.21 / $0.84$0.84
DeepSeek V4 Flash 0731 Uncensored$0.4365 / $1.746$1.75
DeepSeek V4 Pro 0813$0.792 / $1.188$2.18
DeepSeek V4.1 Flash Abliterated$0.8488 / $1.6975$2.55
DeepSeek V4 Pro 0423$0.792 / $2.376$2.77
DeepSeek V3.2 Exp$1.08 / $1.64$2.98
DeepSeek V3.2$1.08 / $1.68$3.00
DeepSeek V4 Pro Uncensored$1.00 / $2.00$3.00
DeepSeek V3$1.00 / $4.00$4.00
DeepSeek V3.1$1.08 / $4.00$4.16
DeepSeek V3.1 Terminus$1.08 / $4.00$4.16
DeepSeek V4.1 Flash Abliterated Fast$1.75 / $3.50$5.25
DeepSeek R1$2.00 / $8.72$8.36

A chat turn is 2,000 input and 500 output tokens. Live from the catalog, ordered by the cost of 1,000 turns.

Worked for DeepSeek V3: 1,000 turns are 2,000,000 input tokens and 500,000 output tokens. At $1.00 input and $4.00 output per million tokens, that is $2.00 + $2.00 = $4.00, paid from prepaid credit with no subscription.

DeepSeek V3 API key · base URL and setup

  1. Create a key in the console at uplinkswitch.ai/console. Credit is prepaid from $10 with no subscription, and the key works at once.
  2. Point your SDK or tool at https://api.uplinkswitch.com/v1, for example with OPENAI_BASE_URL, and send the key as the bearer token.
  3. Send deepseek-v3 as the model. The same key calls every other model in the catalog.
Python · OpenAI SDK
from openai import OpenAI

client = OpenAI(
    base_url="https://api.uplinkswitch.com/v1",
    api_key="sk-uplink-live-...",
)

completion = client.chat.completions.create(
    model="deepseek-v3",
    messages=[{"role": "user", "content": "Summarize this in one line."}],
)
print(completion.choices[0].message.content)
TypeScript · OpenAI SDK
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.uplinkswitch.com/v1',
  apiKey: 'sk-uplink-live-...',
});

const completion = await client.chat.completions.create({
  model: 'deepseek-v3',
  messages: [{ role: 'user', content: 'Summarize this in one line.' }],
});
console.log(completion.choices[0].message.content);
curl
curl https://api.uplinkswitch.com/v1/chat/completions \
  -H "authorization: Bearer sk-uplink-live-..." \
  -H "content-type: application/json" \
  -d '{"model":"deepseek-v3","messages":[{"role":"user","content":"Hello"}]}'

What would it cost? · paste a prompt, priced before you send it

Input is counted at about 4 characters per token.

Cheapest first. A real request is billed on the provider's own token count. Same estimate as JSON: POST https://api.uplinkswitch.com/v1/estimate, no key needed.

DeepSeek V3 questions

DeepSeek V3 is a model from DeepSeek. Through UplinkSwitch it answers the OpenAI SDK on one key as deepseek-v3.

$1.00 per million input tokens and $4.00 per million output tokens.

It is pay as you go: prepaid credit from $10, with no subscription or monthly fee, billed per token. A 1,000-token prompt with a 500-token reply costs about $0.003.

deepseek-v3. Send it as the model through the OpenAI SDK with base URL https://api.uplinkswitch.com/v1.

Create one in the console at uplinkswitch.ai/console: prepaid credit from $10, no subscription, and the key is live at once. The same key calls DeepSeek V3 and every other model in the catalog; send it as the bearer token with base URL https://api.uplinkswitch.com/v1.

Every day each model gets the same test prompt, and the results are public on /proof. DeepSeek V3's latest result: Steady since Oct 20.

Yes: deepseek-v3 on the OpenAI format at https://api.uplinkswitch.com/v1, $1.00 input and $4.00 output per million tokens. No weights to download and no GPU of your own; the key also runs every other model in the catalog.

671B total, 37B active per token, per deepseek-ai/DeepSeek-V3.

128K tokens, per deepseek-ai/DeepSeek-V3.

Its weights are open on Hugging Face (deepseek-ai/DeepSeek-V3); the license is MIT for the code and the DeepSeek Model License for the weights, which allows commercial use.

$0.40 for 100 chat turns of 2,000 input and 500 output tokens each, at $1.00 input and $4.00 output per million tokens, from prepaid credit with no subscription.

V3 is the December 2024 original; V3.2 adds sparse attention and a year of reinforcement learning. DeepSeek V3 is $1.00 / $4.00 per million tokens in / out, $4.00 for 1,000 chat turns; DeepSeek V3.2 is $1.08 / $1.68 per million tokens in / out, $3.00 for 1,000 chat turns. Both run on the same key, so send the same prompt to each.