Skip to content

DeepSeek · Model profile

DeepSeek V3.2 Exp

Intelligence, Performance & Price Analysis

Source: Canonical slug: deepseek-v3-2-0925 · Canonical model registry at build time

Routing verdict

Strong general-purpose model

Suitable for most production tasks, but high-volume or repetitive work should still be compared against cheaper routes.

DeepSeek V3.2 Exp summary metrics

Intelligence
21 scoreDeepSeek V3.2 Exp (Non-reasoning) scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17).
Input Price
$0.28 / 1M tokensDeepSeek V3.2 Exp (Non-reasoning) costs $0.28 per 1M input tokens (very competitive, median: $0.60) and $0.42 per 1M output tokens (very competitive, median: $2.40), based on DeepSeek's API.
Output Price
$0.42 / 1M tokensDeepSeek V3.2 Exp (Non-reasoning) costs $0.28 per 1M input tokens (very competitive, median: $0.60) and $0.42 per 1M output tokens (very competitive, median: $2.40), based on DeepSeek's API.

Assessment

Routing verdict and task fit

DeepSeek V3.2 Exp (Non-reasoning) scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17). DeepSeek V3.2 Exp (Non-reasoning) costs $0.28 per 1M input tokens (very competitive, median: $0.60) and $0.42 per 1M output tokens (very competitive, median: $2.40), based on DeepSeek's API.
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsOptimalIntelligence score 21 supports capable reasoning, but very hard tasks may benefit from higher-tier models.
Cost-sensitive pipelinesOptimalOutput pricing at $0.42 is very competitive for high-volume workloads.

Evidence

Evidence charts

12 published comparison charts for this profile. DeepSeek V3.2 Exp is marked with a diamond and bold label; each chart has a table view and lists its source and retrieval date.

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

17 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 17 models.
  1. Claude Fable 5 (with fallback)40.1
  2. Gemini 3.5 Flash22.7
  3. GPT-5.5 (xhigh)20.1
  4. Grok 4.3 (high)18.3
  5. Kimi K2.66.4
  6. Muse Spark4.1
  7. GLM-5.2 (max)4
  8. MiniMax-M31.4
  9. Nemotron 3 Ultra-0.833
  10. DeepSeek V4 Pro (max)-10
  11. DeepSeek V3 0324-37.6
  12. Mistral Large 3-39.4
  13. DeepSeek V3 (Dec)-41.4
  14. Llama 4 Maverick-41.8
  15. DeepSeek V3.2 Exp-47.1
  16. MiMo-V2-Flash-48.5
  17. gpt-oss-120b (high)-50

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 17 models.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. MiMo-V2-Flash24.7
  12. gpt-oss-120b (high)23.8
  13. DeepSeek V3.2 Exp21.3
  14. Mistral Large 315.9
  15. DeepSeek V3 032415.4
  16. Llama 4 Maverick14.3
  17. DeepSeek V3 (Dec)14.2

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 17 models.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. MiMo-V2-Flash24.7
  12. gpt-oss-120b (high)23.8
  13. DeepSeek V3.2 Exp21.3
  14. Mistral Large 315.9
  15. DeepSeek V3 032415.4
  16. Llama 4 Maverick14.3
  17. DeepSeek V3 (Dec)14.2

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis

10 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 10 models.
  1. Nemotron 3 Ultra83.3
  2. MiMo-V2-Flash52.8
  3. DeepSeek V4 Pro (max)50
  4. DeepSeek V3.2 Exp44.4
  5. GLM-5.2 (max)44.4
  6. gpt-oss-120b (high)38.9
  7. Mistral Large 338.9
  8. Kimi K2.633.3
  9. MiniMax-M333.3
  10. Llama 4 Maverick27.8

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Intelligence

Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis

12 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 12 models.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. DeepSeek V3.2 Exp21.3

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 10 models.
  1. gpt-oss-120b (high)316 t/s
  2. Nemotron 3 Ultra248 t/s
  3. GLM-5.2 (max)216 t/s
  4. Gemini 3.5 Flash192 t/s
  5. Grok 4.3 (high)164 t/s
  6. MiniMax-M395 t/s
  7. GPT-5.5 (xhigh)88 t/s
  8. Kimi K2.676 t/s
  9. DeepSeek V4 Pro (max)72 t/s
  10. Claude Fable 5 (with fallback)71 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

Context Window

Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis

17 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 17 models.
  1. Gemini 3.5 Flash1M
  2. Claude Fable 5 (with fallback)1M
  3. GLM-5.2 (max)1M
  4. DeepSeek V4 Pro (max)1M
  5. Grok 4.3 (high)1M
  6. MiniMax-M31M
  7. Llama 4 Maverick1M
  8. GPT-5.5 (xhigh)922k
  9. Nemotron 3 Ultra262.1k
  10. Muse Spark262.1k
  11. Kimi K2.6256k
  12. MiMo-V2-Flash256k
  13. Mistral Large 3256k
  14. gpt-oss-120b (high)131.1k
  15. DeepSeek V3.2 Exp128k
  16. DeepSeek V3 0324128k
  17. DeepSeek V3 (Dec)128k

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Context window sizing
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis

15 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 15 models.
  1. DeepSeek V3 (Dec)0.00202
  2. Llama 4 Maverick0.00178
  3. DeepSeek V4 Pro (max)0.0063
  4. Mistral Large 30.00455
  5. gpt-oss-120b (high)0.00454
  6. MiniMax-M30.0139
  7. Grok 4.3 (high)0.0101
  8. DeepSeek V3 03240.00334
  9. Nemotron 3 Ultra0.0209
  10. Kimi K2.60.0216
  11. MiMo-V2-Flash0.00812
  12. GLM-5.2 (max)0.0268
  13. Gemini 3.5 Flash0.0884
  14. GPT-5.5 (xhigh)0.107
  15. Claude Fable 5 (with fallback)0.367

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 10 models.
  1. DeepSeek V4 Pro (max)$0.045
  2. gpt-oss-120b (high)$0.061
  3. MiniMax-M3$0.125
  4. Grok 4.3 (high)$0.139
  5. Nemotron 3 Ultra$0.244
  6. Kimi K2.6$0.346
  7. GLM-5.2 (max)$0.370
  8. Gemini 3.5 Flash$0.586
  9. GPT-5.5 (xhigh)$0.863
  10. Claude Fable 5 (with fallback)$2.75

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis

15 models in the published order. DeepSeek V3.2 Exp is not marked in this published set of 15 models.
  1. DeepSeek V3 (Dec)2.9
  2. Llama 4 Maverick4
  3. Mistral Large 38.4
  4. gpt-oss-120b (high)3.5
  5. DeepSeek V3 03245.7
  6. DeepSeek V4 Pro (max)7.2
  7. MiniMax-M314.5
  8. Grok 4.3 (high)12.5
  9. MiMo-V2-Flash11.4
  10. Nemotron 3 Ultra33.5
  11. GLM-5.2 (max)32.9
  12. Kimi K2.625
  13. Gemini 3.5 Flash114
  14. GPT-5.5 (xhigh)131.8
  15. Claude Fable 5 (with fallback)507.6

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis

16 models; segments: Input, Output, Cache Hit. DeepSeek V3.2 Exp is not marked in this published set of 16 models.
  • Input
  • Output
  • Cache Hit
  1. MiMo-V2-Flash$0.400
  2. DeepSeek V3.2 Exp$0.728
  3. gpt-oss-120b (high)$0.900
  4. Llama 4 Maverick$1.47
  5. DeepSeek V3 (Dec)$1.29
  6. DeepSeek V4 Pro (max)$1.31
  7. MiniMax-M3$1.56
  8. Mistral Large 3$2.00
  9. DeepSeek V3 0324$3.64
  10. Nemotron 3 Ultra$3.60
  11. Grok 4.3 (high)$3.95
  12. Kimi K2.6$5.11
  13. GLM-5.2 (max)$6.06
  14. Gemini 3.5 Flash$10.65
  15. GPT-5.5 (xhigh)$35.50
  16. Claude Fable 5 (with fallback)$61.00

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis

12 models; segments: Passive Params, Active Params. DeepSeek V3.2 Exp is not marked in this published set of 12 models.
  • Passive Params
  • Active Params
  1. DeepSeek V4 Pro (max)1,600B
  2. Kimi K2.61,000B
  3. GLM-5.2 (max)753B
  4. DeepSeek V3.2 Exp685B
  5. Mistral Large 3675B
  6. DeepSeek V3 0324671B
  7. DeepSeek V3 (Dec)671B
  8. Nemotron 3 Ultra550B
  9. MiniMax-M3428B
  10. Llama 4 Maverick402B
  11. MiMo-V2-Flash309B
  12. gpt-oss-120b (high)117B

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Architecture scale
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.

Specifications

Architecture and limits

DeepSeek V3.2 Exp technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningNofaq
Input modalitiesDeepSeek V3.2 Exp (Non-reasoning) supports text input.faq
Output modalitiesDeepSeek V3.2 Exp (Non-reasoning) supports text output.faq
Context window130k tokensfaq
Open weights / sourceYes, DeepSeek V3.2 Exp (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersDeepSeek V3.2 Exp (Non-reasoning) has 685 billion parameters (37 billion active).faq
Active parametersDeepSeek V3.2 Exp (Non-reasoning) is a Mixture of Experts (MoE) model with 685 billion total parameters, but only 37 billion active parameters are used during inference.faq
LicenseDeepSeek V3.2 Exp (Non-reasoning) is released under the MIT license. This license allows commercial use.faq
API availabilityYes, DeepSeek V3.2 Exp (Non-reasoning) is available via API through 2 providers.faq

Method

Methodology & Provenance

This page is rendered from the normalized profile and page JSON for DeepSeek V3.2 Exp.

Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.

Questions

Model FAQ and technical disclosures

DeepSeek V3.2 Exp (Non-reasoning) was released on September 29, 2025.

DeepSeek V3.2 Exp (Non-reasoning) was created by DeepSeek.

DeepSeek V3.2 Exp (Non-reasoning) scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 17).

DeepSeek V3.2 Exp (Non-reasoning) costs $0.28 per 1M input tokens (very competitive, median: $0.60) and $0.42 per 1M output tokens (very competitive, median: $2.40), based on DeepSeek's API.

DeepSeek V3.2 Exp (Non-reasoning) costs $0.28 per 1M input tokens and $0.42 per 1M output tokens (based on DeepSeek's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.12 per 1M tokens. Pricing may vary by provider.

No, DeepSeek V3.2 Exp (Non-reasoning) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

DeepSeek V3.2 Exp (Non-reasoning) supports text input.

DeepSeek V3.2 Exp (Non-reasoning) supports text output.

No, DeepSeek V3.2 Exp (Non-reasoning) does not support image input. It can only process text.

No, DeepSeek V3.2 Exp (Non-reasoning) is not multimodal. It only supports text input.

DeepSeek V3.2 Exp (Non-reasoning) has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, DeepSeek V3.2 Exp (Non-reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.

DeepSeek V3.2 Exp (Non-reasoning) has 685 billion parameters (37 billion active).

DeepSeek V3.2 Exp (Non-reasoning) is a Mixture of Experts (MoE) model with 685 billion total parameters, but only 37 billion active parameters are used during inference.

DeepSeek V3.2 Exp (Non-reasoning) is released under the MIT license. This license allows commercial use.

DeepSeek V3.2 Exp (Non-reasoning) achieves a score of 21 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, DeepSeek V3.2 Exp (Non-reasoning) is available via API through 2 providers.

DeepSeek V3.2 Exp (Non-reasoning) is available through 2 API providers.

Models

  • 01DeepSeekOther DeepSeek models in the catalog.
  • 02LeaderboardsCertified metric ranks, when published.