DeepSeek-V3.2-ExpDeepSeek
|
||||||
Related Products
|
||||||
About
Introducing DeepSeek-V3.2-Exp, our latest experimental model built on V3.1-Terminus, debuting DeepSeek Sparse Attention (DSA) for faster and more efficient inference and training on long contexts. DSA enables fine-grained sparse attention with minimal loss in output quality, boosting performance for long-context tasks while reducing compute costs. Benchmarks indicate that V3.2-Exp performs on par with V3.1-Terminus despite these efficiency gains. The model is now live across app, web, and API. Alongside this, the DeepSeek API prices have been cut by over 50% immediately to make access more affordable. For a transitional period, users can still access V3.1-Terminus via a temporary API endpoint until October 15, 2025. DeepSeek welcomes feedback on DSA via its feedback portal. In conjunction with the release, DeepSeek-V3.2-Exp has been open-sourced: the model weights and supporting technology (including key GPU kernels in TileLang and CUDA) are available on Hugging Face.
|
About
Excession API provides hosted access to DeepSeek V4.1 Flash and DeepSeek V4.1 Flash Abliterated through endpoints compatible with OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages. Developers can use the models in their applications with streaming and tool calls, select models by API identifier, and review usage and costs. The API also provides model catalog, API key, account, and balance endpoints, with request rate limits and usage-based billing. It supports developers building AI applications and agents, including authorized security workflows using the abliterated model. Customers can add pay-as-you-go balance or choose a monthly plan with API credit. Both available models have a 1,048,576-token context limit; token rates vary by model, and monthly plan credit expires at renewal.
|
|||||
Platforms Supported
Windows
Supported
Mac
Supported
Linux
Supported
Cloud
Supported
On-Premises
Supported
iPhone
Not Supported
iPad
Not Supported
Android
Not Supported
Chromebook
Not Supported
|
Platforms Supported
Windows
Not Supported
Mac
Not Supported
Linux
Not Supported
Cloud
Supported
On-Premises
Not Supported
iPhone
Not Supported
iPad
Not Supported
Android
Not Supported
Chromebook
Not Supported
|
|||||
Audience
Researchers and AI engineers looking for a solution providing a model that performs well on long-context tasks with reduced compute cost
|
Audience
Developers building AI apps and agents; security teams using Incursion Harness to investigate their systems.
|
|||||
Support
Phone Support
Not Supported
24/7 Live Support
Not Supported
Online
Not Supported
|
Support
Phone Support
Not Supported
24/7 Live Support
Not Supported
Online
Supported
|
|||||
API
Offers API
Supported
|
API
Offers API
Supported
|
|||||
Screenshots and Videos |
Screenshots and Videos |
|||||
Pricing
Free
Open source
Free Version
Supported
Free Trial
Not Supported
|
Pricing
$20/month
Starter costs $20/month and includes $30 in plan credit; Pro is $100/month for $175; Max is $200/month for $400; Ultra is $500/month for $1,250. Each plan covers one person, and unused plan credit expires at renewal. Pay-as-you-go top-ups range from $10 to $1,000; there is no subscription, and the balance does not expire while the account remains open. Per 1 million tokens, DeepSeek V4.1 Flash costs $0.035 for input, $0.005 for cached input, and $0.50 for output. Flash Abliterated costs $0.20 for input, $0.025 for cached input, and $0.80 for output. When all Ultra credit is used, effective rates per 1 million tokens are $0.014/$0.002/$0.20 for Flash and $0.08/$0.01/$0.32 for Flash Abliterated, for input/cached input/output, respectively.
Free Version
Not Supported
Free Trial
Not Supported
|
|||||
Reviews/
|
Reviews/
|
|||||
Training
Documentation
Supported
Webinars
Not Supported
Live Online
Not Supported
In Person
Not Supported
|
Training
Documentation
Supported
Webinars
Not Supported
Live Online
Not Supported
In Person
Not Supported
|
|||||
Company InformationDeepSeek
Founded: 2023
China
deepseek.com
|
Company InformationExcession Labs
excessionlabs.ai/api
|
|||||
Alternatives |
Alternatives |
|||||
|
|
||||||
|
|
|
|||||
|
|
||||||
|
|
||||||
Categories |
Categories |
|||||
Integrations
Anthropic
Not Supported
Claude Code
Not Supported
DeepSeek
Supported
Hugging Face
Supported
OpenAI
Not Supported
OpenCode
Not Supported
|
Integrations
Anthropic
Supported
Claude Code
Supported
DeepSeek
Not Supported
Hugging Face
Not Supported
OpenAI
Supported
OpenCode
Supported
|
|||||
|
|
|