DeepSeek-V3.2: pricing, context window and benchmarks
DeepSeek-V3.2 is a frontier model from DeepSeek. It offers a 1,000,000-token context window, costs $5.88 per million input tokens and $24.70 per million output tokens, and averages 84.0 across the index's benchmark suite.
Pricing
DeepSeek-V3.2 costs $5.88 per million input tokens and $24.70 per million output tokens. A blended 3:1 input-to-output mix works out to about $10.59 per million tokens. Cached input is listed at $0.50 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $5.88 | per 1M tokens |
| Output | $24.70 | per 1M tokens |
| Cached input | $0.50 | per 1M tokens |
| Blended 3:1 | $10.59 | estimate per 1M tokens |
| 1M in + 1M out | $30.58 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 86.9 | 37 | 94th | |
| GPQA Diamond | 79.7 | 63 | 90th | |
| SWE-bench Verified | 74.4 | 47 | 92th | |
| HumanEval | 89.7 | 53 | 91th | |
| MATH-500 | 89.7 | 52 | 91th | |
| AIME 2025 | 79.6 | 93 | 85th | |
| LiveCodeBench | 80.4 | 40 | 93th | |
| IFEval | 89.4 | 77 | 87th | |
| TruthfulQA | 84.5 | 73 | 88th | |
| HellaSwag | 90.2 | 78 | 87th | |
| MMMU | 79.9 | 57 | 91th | |
| Arena Elo | 1455.6 | 62 | 90th |
What DeepSeek-V3.2 is good for
DeepSeek-V3.2 sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $5.88 per million input tokens and a 1,000,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.
Strengths
- Top-tier average benchmark score (84.0), placing it #57 of 600.
- Large 1,000,000-token context window for long documents and agent traces.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
Trade-offs
- Cost can climb quickly on long outputs and high-volume traffic.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Gemini 3 Flash Turbo | Google DeepMind | 200000 | $17.07 | 84.1 |
| DeepSeek-V3.2 | DeepSeek | 1000000 | $5.88 | 84.0 |
| Gemini 2.5 Pro Turbo | Google DeepMind | 1000000 | $16.32 | 83.9 |
| Claude Haiku 4.5 Max | Anthropic | 200000 | $15.98 | 83.8 |
| GPT-4.1 Reasoning | OpenAI | 1000000 | $16.97 | 83.8 |
| Grok 3 Pro | xAI | 400000 | $16.61 | 83.8 |
FAQ
How much does DeepSeek-V3.2 cost?
DeepSeek-V3.2 is listed at $5.88 per million input tokens and $24.70 per million output tokens. Cached input is $0.50 per million tokens. A blended 3:1 input-to-output mix is about $10.59 per million tokens.
What is the context window of DeepSeek-V3.2?
DeepSeek-V3.2 has a context window of 1,000,000 tokens and can emit up to 8,192 output tokens per request in this index.
Is DeepSeek-V3.2 open weights?
Yes. DeepSeek-V3.2 is listed as an open-weight model under a open weights licence.
How does DeepSeek-V3.2 score on benchmarks?
DeepSeek-V3.2 averages 84.0 across the scored benchmarks in this index, ranking #57 of 600 models. Its strongest relative result is Arena Elo at 1455.6.
What is DeepSeek-V3.2 best for?
DeepSeek-V3.2 sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $5.88 per million input tokens and a 1,000,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.