DeepSeek V3.2 Exp
deepseek/deepseek-v3.2-expDeepSeek chat model for instruction following, coding, and analysis
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...
Context
164K
163,840 tokens
Max output
147K
147,456 tokens
Input
$0.27
per 1M tokens
Output
$0.41
per 1M tokens
Providers
5
5 listings
Intelligence
16.6
Intelligence Index
Pricing
USD per 1M tokens · listed by Kilo Gateway
| Token type | Standard |
|---|---|
| Input | $0.27 |
| Output | $0.41 |
Lowest input price across providers: $0.22.
Capabilities
ReasoningYes
Effort: none, high
Tool callYes
Structured outputYes
TemperatureYes
AttachmentsNo
Interleaved reasoningNo
Open weightsNo
ModeratedNo
Input
Text
Output
Text
Benchmarks
Published scores and Design Arena Elo
- Intelligence16.6#117 of 208
- CodingNot scored—
- AgenticNot scored—
- Arena1,140#76 of 128
Rankings
Position on each OneAtlas leaderboard
Providers · 5
5 listings
| Provider | Model ID | Context | Input limit | Output | Input $ | Output $ | Reasoning $ | Cache read | Cache write | Audio in | Audio out | Reasoning | Tools | Structured | Temp. | Updated |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
 Alibaba (China) | deepseek-v3-2-exp | 131K | — | 66K | $0.29 | $0.43 | — | — | — | — | — | No | Yes | Unknown | Yes | Jan 1, 2025 |
 Kilo Gateway | deepseek/deepseek-v3.2-exp | 164K | — | 147K | $0.27 | $0.41 | — | — | — | — | — | Yes | Yes | Yes | Yes | Sep 29, 2025 |
 Meganova | deepseek-ai/DeepSeek-V3.2-Exp | 164K | — | 164K | $0.26 | $0.38 | — | $0.13 | — | — | — | No | Yes | Yes | Yes | Oct 10, 2025 |
 NanoGPT | deepseek-ai/deepseek-v3.2-exp | 164K | 164K | 66K | $0.28 | $0.42 | — | $0.14 | — | — | — | No | No | No | Unknown | Sep 29, 2025 |
 ZenMux | deepseek/deepseek-v3.2-exp | 163K | — | 64K | $0.22 | $0.33 | — | — | — | — | — | Yes | Yes | Unknown | Yes | Sep 29, 2025 |