DeepSeek
DeepSeek V3
DeepSeek chat model for instruction following, coding, and analysis
Model facts
- Context window
- 164K tokens
- Maximum output
- 8K tokens
- Cheapest paid input
- $0.32 per 1M tokens
- Cheapest paid output
- $0.89 per 1M tokens
- Open weights
- yes
- Providers
- 12
Capabilities and modalities
tool calling, structured output, temperature control, open weights, text.
Local hardware estimate
At 4K context: Q4_K_M 419.9 GiB, Q8_0 738.7 GiB, F16 1336.4 GiB working memory.
- Parameters
- 684.5 billion
- Architecture
- deepseekv3
- Model license
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| Dec '24 | 14.2 | 23.0 | 26.0 | not measured |
Observed price history
- 2026-08-24: $0.32 input / $0.89 output per 1M tokens
API providers
- iFlow — model id deepseek-v3; free input / free output per 1M tokens; provider documentation
- Deep Infra — model id deepseek-ai/DeepSeek-V3; $0.32 input / $0.89 output per 1M tokens; provider documentation
- Eden AI — model id deepinfra/deepseek-ai/DeepSeek-V3; $0.32 input / $0.89 output per 1M tokens; provider documentation
- D.Run (China) — model id public/deepseek-v3; $0.28 input / $1.1 output per 1M tokens; provider documentation
- Pioneer — model id deepseek-ai/DeepSeek-V3; $0.27 input / $1.12 output per 1M tokens; provider documentation
- Alibaba (China) — model id deepseek-v3; $0.287 input / $1.15 output per 1M tokens; provider documentation
- Hugging Face — model id deepseek-ai/DeepSeek-V3; $0.4 input / $1.3 output per 1M tokens; provider documentation
- Helicone — model id deepseek-v3; $0.56 input / $1.68 output per 1M tokens; provider documentation
- Merge Gateway — model id deepseek/deepseek-v3; $0.58 input / $1.68 output per 1M tokens; provider documentation
- Together AI — model id deepseek-ai/DeepSeek-V3; $1.25 input / $1.25 output per 1M tokens; provider documentation
- DigitalOcean — model id deepseek-v3; not published input / not published output per 1M tokens; provider documentation
- Qiniu — model id deepseek-v3; not published input / not published output per 1M tokens; provider documentation