deepseek logo
deepseek

DeepSeek: DeepSeek V4.1 Flash

Descrizione della fonte (inglese)

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...

Panoramica

Specifiche del modello

Contesto
1.048.576 tokens
Output massimo
384.000 tokens
Architettura
text+image->text
Tokenizer
DeepSeek
Limite di conoscenza
Non indicato
Moderato
No
OPENROUTER

Prezzi completi

Tariffe sincronizzate da OpenRouter, per milione di token.

Input
$0.15
per milione di token
Output
$0.6
per milione di token
Lettura cache
$0.003
per milione di token
API

Configurazione del modello

Ragionamento

Ragionamento obbligatorio
No
Parametri predefiniti
high
Capacità e modalità
max, high, low
API

Avvio rapido

Imposta OPENROUTER_API_KEY localmente. Python richiede requests; JavaScript viene eseguito in Node.js. Conserva la chiave sul server.

Documentazione API

curl --fail-with-body https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek/deepseek-v4.1-flash","messages":[{"role":"user","content":"Hello!"}]}'
Capacità e modalità

InputOutput

Input
textimage
Output
text
Ragionamento

· max · high · low

API

Parametri API supportati

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Provider disponibili

18 provider

Verificato: 15 settembre 2026

Provider live su OpenRouter

Disponibilità, latenza, throughput e routing cambiano continuamente. Consulta la fonte per i dati correnti.

OpenRouter

Relace

fp4
Contesto
1.0M
Output massimo
944K
Input
$0.15
Output
$0.6
Lettura cache
$0.015
Scrittura cache
Non indicato

DeepSeek

Non indicato
Contesto
1.0M
Output massimo
384K
Input
$0.15
Output
$0.6
Lettura cache
$0.003
Scrittura cache
Non indicato

DeepInfra

fp8
Contesto
1.0M
Output massimo
131K
Input
$0.2
Output
$0.6
Lettura cache
$0.006
Scrittura cache
Non indicato

Fireworks

Non indicato
Contesto
1.0M
Output massimo
944K
Input
$0.22
Output
$0.66
Lettura cache
$0.007
Scrittura cache
Non indicato

Morph

fp8
Contesto
1.0M
Output massimo
944K
Input
$0.225
Output
$0.9
Lettura cache
$0.0225
Scrittura cache
Non indicato

Io Net

fp8
Contesto
262K
Output massimo
131K
Input
$0.285
Output
$1.14
Lettura cache
$0.00285
Scrittura cache
Non indicato

Reka

fp4
Contesto
262K
Output massimo
131K
Input
$0.29
Output
$1.16
Lettura cache
$0.029
Scrittura cache
Non indicato

Alibaba

Non indicato
Contesto
1M
Output massimo
393K
Input
$0.3
Output
$1.2
Lettura cache
$0.03
Scrittura cache
Non indicato

Together

Non indicato
Contesto
1.0M
Output massimo
944K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

SiliconFlow

fp8
Contesto
1.0M
Output massimo
393K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

Modal

Non indicato
Contesto
1.0M
Output massimo
944K
Input
$0.3
Output
$1.2
Lettura cache
$0.03
Scrittura cache
Non indicato

Wafer

Non indicato
Contesto
1.0M
Output massimo
944K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

BaseTen

fp8
Contesto
1.0M
Output massimo
33K
Input
$0.3
Output
$1.2
Lettura cache
$0.03
Scrittura cache
Non indicato

Parasail

fp8
Contesto
1.0M
Output massimo
944K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

GMICloud

fp8
Contesto
1.0M
Output massimo
944K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

Novita

fp8
Contesto
1.0M
Output massimo
393K
Input
$0.3
Output
$1.2
Lettura cache
$0.006
Scrittura cache
Non indicato

Phala

Non indicato
Contesto
1.0M
Output massimo
393K
Input
$0.345
Output
$1.38
Lettura cache
$0.0069
Scrittura cache
Non indicato

Venice

fp8
Contesto
1M
Output massimo
131K
Input
$0.375
Output
$1.5
Lettura cache
$0.0075
Scrittura cache
Non indicato
deepseek

4 modelli

Tutti i modelli