Skip to content
Not available in this workspace
OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Collections
  • Providers
  • Pricing
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Support
  • Works With OR
  • Data

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for deepseek

DeepSeek: DeepSeek V3.2 Exp

deepseek/deepseek-v3.2-exp

Model weights
Compare

DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism designed to improve training and inference efficiency in long-context scenarios while maintaining output quality. Users can control the reasoning behaviour with the reasoning enabled boolean. Learn more in our docs(opens in new tab)

The model was trained under conditions aligned with V3.1-Terminus to enable direct comparison. Benchmarking shows performance roughly on par with V3.1 across reasoning, coding, and agentic tool-use tasks, with minor tradeoffs and gains depending on the domain. This release focuses on validating architectural optimizations for extended context lengths rather than advancing raw task accuracy, making it primarily a research-oriented model for exploring efficient transformer designs.

Modalities

In / Out Price

$0.27 / $0.41per 1M

Context

164K

Released

Sep 29, 2025

Knowledge Cutoff

Jul 2025

Compare
ProvidersPricingPerformanceUptimeBenchmarksAppsActivityFAQ

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Benchmarks

Scores on standardized evaluations. Higher percentages are better — and rank percentile shows where this model lands among all models on OpenRouter.

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Activity

Token volume and request traffic to this model over time.

Quick Start

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

About DeepSeek: DeepSeek V3.2 Exp

OpenRouter makes DeepSeek: DeepSeek V3.2 Exp available through a unified, OpenAI-compatible API using the model ID deepseek/deepseek-v3.2-exp. Requests can be routed across 3 providers, including NovitaAI, SiliconFlow and AtlasCloud, with automatic failover when an endpoint is unavailable.

DeepSeek: DeepSeek V3.2 Exp accepts text and returns text. It has a 163,840-token context window and a maximum output of 65,536 tokens.

On OpenRouter, DeepSeek: DeepSeek V3.2 Exp costs $0.27/M input tokens and $0.41/M output tokens. It was released on September 29, 2025; its knowledge cutoff is July 31, 2025.

More models from DeepSeek

  • DeepSeek V4 Pro 0813
  • DeepSeek V4 Flash 0731
  • DeepSeek V4 Pro 0423

Frequently asked questions

DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism designed to improve training and inference efficiency in long-context scenarios while maintaining output quality.

DeepSeek V3.2 Exp costs $0.27/M input tokens and $0.41/M output tokens.

DeepSeek V3.2 Exp has a 163,840 token context window. It supports up to 65,536 completion tokens.

Yes. DeepSeek V3.2 Exp accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

DeepSeek V3.2 Exp is served by 3 providers on OpenRouter: NovitaAI, SiliconFlow and AtlasCloud. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

DeepSeek V3.2 Exp was released on September 29, 2025. Its knowledge cutoff is July 31, 2025.

$0.27$0.41--1.85s26 tps
99.91%
$0.27$0.41--2.02s25 tps
99.84%
$0.27$0.41$0.271.75s29 tps
99.49%

Throughput

29tok/s

P50, best across providers

Latency

1.75s

P50, best provider

AutoExacto Benchmarks
GPQA DiamondTAU-Bench
auto-routing
81.9%72.2%
AtlasCloud
82.9%70.7%
SiliconFlow
82.1%69.3%
NovitaAI
82.1%67.3%
Uptime (3d)

100.00%

Availability (3d)

99.99%

Availability over the last 3 days

Last 72 hours
Availability 99.99%
3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability
99.99%
Without Routing
96.72%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and customization options.

1.
Favicon for https://janitorai.com/
Janitor AI
Janitor AI is a chatbot platform where users create and chat with custom AI characters for interactive roleplay, storytelling, and immersive fiction.
13.7Btokens
2.
Favicon for https://chub.ai/
Chub AI
GenAI for everyone
3.26Btokens
3.
Favicon for https://sillytavern.app/
SillyTavern
SillyTavern is the LLM frontend for power users, a chat interface that connects to any model and gives deep control through character creation, roleplay, and prompt customization.
3.13Btokens
4.
Favicon for https://revla.ai/platform/3
revla:3
new
2.62Btokens
5.
Favicon for https://revla.ai/platform/13
revla:13
new
1.66Btokens