Sąskaita sudaroma pagal faktiškai panaudotus žetonus; neišleista rezervo dalis grąžinama.
Naujas čia?
10 nemokami kreditai (0,10 USD), kai užsiregistruojate naudodami Google
Galima naudoti 24 valandų po registracijos, iki 5 paleidimų per dieną ir daugiausia 2 kreditų per paleidimą. Kiti prisijungimo būdai pradedami be kreditų. Pakanka 66 šio modelio paleidimams.
02
Apie DeepSeek V4 Flash
TrumpaiNuo 2026 m. rugsėjo 23 d.
DeepSeek V4 Flash yra DeepSeek modelis iš Tekstas ir pokalbis kategorijos. Per Railwail DeepSeek V4 Flash kainuoja 0,36 USD už 1M įvesties žetonų ir 1,44 USD už 1M išvesties žetonų. Konteksto langas saugo 1 048 575 žetonų, o vienas atsakymas gali būti iki 384 000 žetonų ilgio. Naujesnė versija: 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.
Fonas
Apie DeepSeek AI
Įkurta 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.
Parametrai
284B total / 13B active per token
Kontekstas
1 048 575 žetonų
Galimybės
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.
Mokymas ir licencija
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.
Licencija: Open weights under a permissive license that allows commercial use. Hosted API access via deepseek.com.
Saugumo testai: DeepSeek publishes model cards but provides limited external red-teaming. Safety filters are lighter than Western frontier labs; deployers are responsible for downstream alignment.
Žinomos ribotybės
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 yra DeepSeek modelis iš Tekstas ir pokalbis kategorijos. Railwail platformoje galite jį iškviesti naudodami API raktą per Railwail API.
Kiek kainuoja DeepSeek V4 Flash Railwail?
Per Railwail DeepSeek V4 Flash kainuoja 0,36 USD už 1M įvesties žetonų ir 1,44 USD už 1M išvesties žetonų. Jums sąskaitoje pateikiama suma, kurią iš tikrųjų sunaudoja kiekvienas užklausa. Naudojimas apmokamas iš išanksto nupirktų kreditų; 1 kreditas = 0,01 USD.
Koks yra DeepSeek V4 Flash konteksto langas?
DeepSeek V4 Flash konteksto langas saugo 1 048 575 žetonų. Vienas atsakymas gali būti iki 384 000 žetonų ilgio.
Koks yra DeepSeek V4 Flash greitis?
DeepSeek V4 Flash Railwail platformoje dar nėra pakankamai išmatuotų vykdymų, kad būtų galima nurodyti vykdymo laiką. Jis priklauso nuo įvesties, nustatymų ir teikėjo apkrovos.
Ar DeepSeek V4 Flash geriau nei DeepSeek V4.1 Flash?
Tai priklauso nuo užduoties. DeepSeek V4 Flash (DeepSeek) ir DeepSeek V4.1 Flash (DeepSeek) yra abu modeliai iš Tekstas ir pokalbis kategorijos. Palyginimo puslapyje jų kainos ir specifikacijos rodomos viena šalia kitos.
Sukurkite Railwail API raktą ir siųskite savo užklausą su modelio ID deepseek-v4-flash. Kodo pavyzdžiai curl, Python ir JavaScript yra šio puslapio API skyriuje.
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.
