DeepSeek V3.1

reasoningdeepseek

Hybrid reasoning MoE model with switchable thinking and non-thinking modes.

Context
131K
Max output
7K
Input price
$3/1M tokens
Output price
$4.5/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Fine-tuning

Details

Model IDDeepSeek-V3.1
Provider SambaNova Cloud
Creator DeepSeek
Familydeepseek-v3
Licensedeepseek
Statusactive
Typereasoning
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters—
Architecture—
Context window131K tokens
Max input—
Max output7K tokens
Reasoning tokensYes
Knowledge cutoff—
Release date—
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.sambanova.ai/v1/chat/completions

Pricing

Input
$3
Output
$4.5
Cache write
—
Cache read
—
Batch in
—
Batch out
—

Family Comparison: deepseek-v3

ModelContextMax outPricing
DeepSeek V3.1 (continuous batching)33K33K$0.15$0.75
DeepSeek V3.1131K7K$3$4.5
DeepSeek V3.233K7K$3$4.5

API

GET/v1/models/sambanova/DeepSeek-V3.1

DeepSeek V3.1

DeepSeek-V3.1

Hybrid reasoning MoE model with switchable thinking and non-thinking modes.

Changes · 5 entries
DeepSeek-V3.1update2f25bc2Oct 29, 2026, 06:34 AM
max_output_tokens131072→7168
DeepSeek-V3.1updateb2f45bfOct 29, 2026, 03:29 AM
max_output_tokens7168→131072
DeepSeek-V3.1update20c73acJul 17, 2026, 02:34 AM
max_output_tokens131072→7168
DeepSeek-V3.1update53ef452Jul 9, 2026, 02:32 AM
max_output_tokens7168→131072
DeepSeek-V3.1create3e56cf1Apr 26, 2026, 07:23 PM