Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. All models overview
  2. Text
  3. DeepSeek

DeepSeek V4 Pro 0813

DeepSeek V4 Pro 0813 is the general-availability release of DeepSeek V4 Pro, built for coding, tool use, cybersecurity, automation, and long-horizon agent workflows. DeepSeek reports strong gains over the preview and scores above Opus 4.8 on Terminal Bench 2.1, Cybergym, DeepSWE, and AutomationBench. It supports a 1M-token context window.

  • Reasoning
  • Tool Calling
  • Structured Output

Added Aug 12, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.66
Output
$2.50
Cache read
$0.04
Compare provider prices

Specifications

Context window
1M
Max output
384K
Parameters
1.6T / 49B
Total / active

Benchmarks

Leaderboard results

BenchmarkScoreRank on NanoGPT
Epoch AI Benchmarking Hub

Epoch AI data, CC BY 4.0 · Updated Nov 21, 2026

Epoch Capabilities Index

156.0

#39 of 114

FrontierMath v2 (Tiers 1–3)

64.6%

#40 of 84

GPQA Diamond

91.7%

#30 of 119

SimpleQA Verified

52.9%

#29 of 84

FrontierMath v2 (Tier 4)

26.8%

#40 of 68

OTIS Mock AIME 2024–2025

98.6%

#28 of 121

MirrorCode

33.3%

#31 of 41

Chess Puzzles

46.0%

#26 of 83

Mystery Game Puzzles

43.0%

#21 of 80

Filtered and ranked to models on NanoGPT.

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.66$2.50$0.04N/AN/A
Auto
Auto (Anonymized)
$0.66$2.50$0.04N/AN/A
Relace
Relace
FP4
UNK
$0.105$8.40$0.105N/AN/A
Baidu
Baidu
90% off
FP8
CN
$1.39$0.139$4.16$0.416$0.0441$0.0044N/AN/A
Wafer
Wafer
MXFP4
US
$0.294$4.16$0.0462N/AN/A
DeepSeek
DeepSeek
50% off
CN
$1.39$0.693$4.16$2.08$0.0462$0.0231N/AN/A
StreamLake
StreamLake
50% off
CN
$1.39$0.693$4.16$2.08$0.0462$0.0231N/AN/A
Kitani
Kitani
GLB
$1.26$3.99$0.0462N/AN/A
Novita
Novita
FP8
US
$1.32$3.95$0.0439N/AN/A
DeepInfra
DeepInfra
FP8
US
$1.36$2.73$0.105N/AN/A
CoreWeave
CoreWeave
FP8
US
$1.38$4.16$0.0462N/AN/A
BaseTen
BaseTen
FP4
US
$1.39$4.16$0.139N/AN/A
Inworld
Inworld
US
$1.39$4.16$0.0462N/AN/A
Nebius
Nebius
FP8
GLB
$1.39$4.16$0.042N/AN/A
Ollama
Ollama
UNK
$1.39$4.16$0.0462N/AN/A
Parasail
Parasail
FP8
US
$1.39$4.16$0.0462N/AN/A
SiliconFlow
SiliconFlow
FP8
SG
$1.39$4.16$0.0462N/AN/A
Together
Together
US
$1.39$4.16$0.136N/AN/A
Venice
Venice
US
$1.73$5.20$0.173N/AN/A
TensorX
TensorX
EU
$2.10$4.20$0.525N/AN/A
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclearMay train on prompts
Batch API
ProviderInputOutputCached input
Auto
Autodeepseek/deepseek-v4-pro-0813
$0.66/M$2.50/M$0.04/M

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare DeepSeek V4 Pro 0813 with similar models from the same provider or model family.

DeepSeek V4 Pro 0813 Thinking

deepseek/deepseek-v4-pro-0813:thinking

DeepSeek V4 Pro 0813 Thinking enables reasoning by default on the general-availability release of DeepSeek V4 Pro, built for coding, tool use, cybersecurity, automation, and long-horizon agent workflows. DeepSeek reports strong gains over the preview and scores above Opus 4.8 on Terminal Bench 2.1, Cybergym, DeepSWE, and AutomationBench. It supports a 1M-token context window.

DeepSeek V4 Pro (April preview)

deepseek/deepseek-v4-pro-20260423

DeepSeek V4 Pro April preview (2026-04-23), the first V4 Pro checkpoint: a large-scale Mixture-of-Experts model with a 1M-token context window for reasoning, coding, long-horizon agent workflows, knowledge, math and software engineering. Plain DeepSeek V4 Pro now means V4 Pro 0813.

DeepSeek V4 Pro (April preview, Thinking)

deepseek/deepseek-v4-pro-20260423:thinking

DeepSeek V4 Pro April preview (2026-04-23) with chain-of-thought enabled: a large-scale Mixture-of-Experts model with a 1M-token context window. Plain DeepSeek V4 Pro now means V4 Pro 0813.

DeepSeek V4.1 Flash 1017 Abliterated

deepseek/deepseek-v4.1-flash-1017-abliterated

An abliterated (refusal-removed) version of DeepSeek V4.1 Flash 1017. Text only, with tool calling, optional reasoning and a 1M-token context window. Its behavior may differ from DeepSeek V4.1 Flash 1017. Served by a single host; it may be unavailable while that host is down.

DeepSeek V4.1 Flash 1017 Abliterated Thinking

deepseek/deepseek-v4.1-flash-1017-abliterated:thinking

An abliterated (refusal-removed) version of DeepSeek V4.1 Flash 1017. Text only, with tool calling, optional reasoning and a 1M-token context window. Its behavior may differ from DeepSeek V4.1 Flash 1017. Served by a single host; it may be unavailable while that host is down.

DeepSeek V4.1 Flash Abliterated

deepseek/deepseek-v4.1-flash-abliterated

An experimental abliterated (refusal-removed) version of DeepSeek V4.1 Flash, hosted by UncensuredAI. It always reasons, with low, high or max effort, and supports tool calling. Text only. Its behavior and response quality may differ from DeepSeek V4.1 Flash.