Наплаћују се стварно потрошени токени; неискоришћени део резервације се враћа.
Нов си овде?
10 бесплатних кредита (0,10 US$) када се пријавиш са Google
Употребљиво 24 сати после пријаве, до 5 покрета дневно и највише 2 кредита по покрету. Други начини пријаве почињу без кредита. Довољно за 66 покретања овог модела.
02
О DeepSeek V4 Flash
УкраткоСтање: 23. септембар 2026.
DeepSeek V4 Flash је модел од DeepSeek из категорије Текст и разговор. На Railwail-у, DeepSeek V4 Flash кошта 0,36 US$ по 1М улазних токена и 1,44 US$ по 1М излазних токена. Контекстни прозор садржи 1.048.575 токена, а један одговор може бити дугачак до 384.000 токена. Новија верзија: DeepSeek V4.1 Flash.
DeepSeek-V4-Flash is the cost-efficient sibling of V4-Pro, released April 2026 as part of the V4 Preview. 284B total / 13B active MoE parameters with the same 1M-token context window. Designed for high-throughput agentic loops, RAG and batch tasks where latency and cost matter more than raw capability. Recommended for production agents, classification at scale, large-scale data extraction.
Позадина
О DeepSeek AI
Основано 2023 · Hangzhou, China
DeepSeek AI is a Chinese AI research lab founded in 2023 by Liang Wenfeng, founder of the High-Flyer quantitative hedge fund. The lab is funded primarily by High-Flyer's profits. Its mission is open frontier AI, with all flagship models released with open weights. Major releases include DeepSeek LLM (2023), DeepSeek-V2 (May 2024), DeepSeek-V3 (December 2024), DeepSeek-R1 (January 2026), DeepSeek V3.1 (early 2026) and the DeepSeek V4 family (April 24, 2026), comprising V4-Pro and V4-Flash. DeepSeek is credited with popularising large-scale Reinforcement Learning from Verifiable Rewards and consistently tops open-weights leaderboards.
DeepSeek-V4-Flash was released April 24, 2026 as the efficiency-optimized sibling of V4-Pro. It is a Sparse MoE Transformer with 284B total parameters and 13B activated per token, retaining the full 1M-token native context window and 384K-token max output of the Pro variant at significantly lower inference cost. The model uses the same DeepSeek architectural stack: Multi-head Latent Attention (MLA), DeepSeekMoE with fine-grained expert specialization and shared experts, and FP8 mixed-precision training. Post-training combined supervised fine-tuning, RLVR on math/code/tool-use trajectories, and heavy distillation from the V4-Pro teacher model. V4 Flash is published with open weights under a permissive license and is designed for production-scale RAG, agentic loops and high-throughput workloads. At $0.112 input / $0.224 output per million tokens it undercuts every Western frontier model by an order of magnitude.
Параметри
284B total / 13B active per token
Контекст
1.048.575 токена
Могућности
1M token native context window with 384K max output
284B MoE / 13B active parameters
Ultra-low pricing ($0.112 / $0.224 per million tokens)
Distilled from DeepSeek V4-Pro teacher model
FP8-trained for compute efficiency
Multi-head Latent Attention for memory-efficient long context
Function calling and structured JSON output
Strong on math, STEM and coding for its size
Available via DeepSeek API, OpenRouter, Together and self-hosted with vLLM/SGLang
Open weights under a permissive license
Best for: production agents, RAG pipelines, high-throughput data extraction, on-premise inference under tight cost budgets.
Обука и лиценца
Pretrained on the same multi-trillion-token mixture as V4-Pro. Post-training combines supervised fine-tuning, RLVR and distillation from the V4-Pro teacher model. Knowledge cutoff approximately early 2026.
Лиценца: Open weights under a permissive license that allows commercial use. Hosted API access via deepseek.com.
Тестирање безбедности: DeepSeek publishes model cards but provides limited external red-teaming. Safety filters are lighter than Western frontier labs; deployers are responsible for downstream alignment.
Позната ограничења
Below V4-Pro on the hardest reasoning and coding benchmarks
Light built-in safety alignment relative to Western frontier models
No native vision or audio input (text-only)
Older deepseek-chat / deepseek-reasoner endpoints will be deprecated July 24, 2026
curl https://railwail.com/api/v1/chat/completions \
-H "Authorization: Bearer $RAILWAIL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-flash",
"messages": [
{
"role": "user",
"content": "Explain what a vector database is in two sentences."
}
],
"max_tokens": 1024
}'
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["RAILWAIL_API_KEY"],
base_url="https://railwail.com/api/v1",
)
completion = client.chat.completions.create(
model="deepseek-v4-flash",
messages=[
{
"role": "user",
"content": "Explain what a vector database is in two sentences.",
},
],
max_tokens=1024,
)
print(completion.choices[0].message.content)
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.RAILWAIL_API_KEY,
baseURL: "https://railwail.com/api/v1",
});
const completion = await client.chat.completions.create({
model: "deepseek-v4-flash",
messages: [
{
role: "user",
content: "Explain what a vector database is in two sentences."
}
],
max_tokens: 1024
});
console.log(completion.choices[0].message.content);
// npm install railwail
import railwail from "railwail";
const rw = railwail(process.env.RAILWAIL_API_KEY);
const res = await rw.chat("deepseek-v4-flash", [
{ role: "user", content: "Explain what a vector database is in two sentences." },
], { max_tokens: 1024 });
console.log(res.choices[0].message.content);
DeepSeek V4 Flash је модел од DeepSeek у категорији Текст и разговор. На Railwail-у га можете позвати са API кључем кроз Railwail API.
Колико кошта DeepSeek V4 Flash на Railwail-у?
На Railwail-у, DeepSeek V4 Flash кошта 0,36 US$ по 1М улазних токена и 1,44 US$ по 1М излазних токена. Наплаћује се оно што свака захтева заиста користи. Коришћење се плаћа из унапред припремљених кредита; 1 кредит је једнак 0,01 US$.
Колики је контекстни прозор од DeepSeek V4 Flash?
Контекстни прозор од DeepSeek V4 Flash садржи 1.048.575 токена. Један одговор може бити дугачак до 384.000 токена.
Колико је брз DeepSeek V4 Flash?
За DeepSeek V4 Flash на Railwail-у још нема довољно измерених покретања да би се наводило време извршавања. Зависи од уноса, поставки и оптерећења код пружаоца услуге.
Да ли је DeepSeek V4 Flash бољи од DeepSeek V4.1 Flash?
То зависи од задатка. DeepSeek V4 Flash (DeepSeek) и DeepSeek V4.1 Flash (DeepSeek) су оба модела у категорији Текст и разговор. Страница за поређење приказује њихове цене и спецификације упоредо.
Направите Railwail API кључ и пошаљите ваш захтев са ID модела deepseek-v4-flash. Примери кода за curl, Python и JavaScript налазе се у API одељку ове странице.
Anthropic's model for the most demanding reasoning and long-horizon agentic work. 1M-token context window, up to 128K output tokens, adaptive thinking that is always on.
The most capable model of Anthropic's Opus 4 series. State of the art on long-horizon agentic work, coding and knowledge tasks, with a 1M-token context window at standard pricing.
Anthropic's current Opus model for long-running agentic coding and knowledge work. 1M-token context window, up to 128K output tokens, adaptive thinking that is always on.
