DeepSeek-V4-Pro (2026-08-13)
- In Input modalities: Text Out Output modalities: Text
- Reasoning· low, high, max
- Reasoning can be off
- Tool use
- Structured output
- Open weights
Prices
8 offerings · prices as listeddeepseek-v4-pro-0813| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Peak hours Standard | $1.272 | $3.816 |
| Off-peak hours | $0.636 | $1.908 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Nov 2, 2026)
deepseek-v4-pro-0813| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Peak hours Standard | $1.32 | $3.96 |
| Off-peak hours | $0.66 | $1.98 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Nov 2, 2026)
deepseek-v4-pro-0813| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard | ¥9.00 | ¥27.00 |
| Off-peak hours | ¥4.50 | ¥13.50 |
Prices from 阿里云百炼：模型价格 (opens in a new tab) (checked Nov 2, 2026)
vanchin/deepseek-v4-pro-0813| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard | ¥9.00 | ¥27.00 |
Prices from 阿里云百炼：模型价格 (opens in a new tab) (checked Nov 2, 2026)
deepseek-v4-pro-0813| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Peak hours no longer listed since Oct 23, 2026 | $1.272 | $3.816 |
| Off-peak hours no longer listed since Oct 23, 2026 | $0.636 | $1.908 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Nov 2, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Licence
- MIT
- Family
- deepseek-pro
- Model ID
Not stated on DeepSeek’s pages: max input, max thinking budget, built-in tools, knowledge cutoff, aliases.
Run locally
Weights: deepseek-ai/DeepSeek-V4-Pro-0813- Parameters 1.65T
- Active 127.3B (estimated) · 384 experts, 6 per token
- Native precision FP8
- Layers 61
- KV cache per token 122 KB
- Config positions 1024K (config.json max_position_embeddings)
- API context 977K (maker’s API)
- Needs
- ≈ 858.4 GB (weights 816.6 + KV cache 1.0 + overhead 40.8)
- Usable
- 21.6 GB (90% of its memory)
- Try
- it’s too large for this hardware; see the smallest that runs it below
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| FP8 (native)maker’s file | 831.4 GB | ≈ 874.0 | ✕Doesn’t fit | — |
| Q6_K | 1262.4 GB | ≈ 1326.5 | ✕Doesn’t fit | — |
| Q5_K_M | 1093.3 GB | ≈ 1148.9 | ✕Doesn’t fit | — |
| Q5_0 | 1056.8 GB | ≈ 1110.6 | ✕Doesn’t fit | — |
| Q4_K_M | 933.8 GB | ≈ 981.5 | ✕Doesn’t fit | — |
| Q4_0 | 864.6 GB | ≈ 908.8 | ✕Doesn’t fit | — |
| IQ4_XS | 816.6 GB | ≈ 858.4 | ✕Doesn’t fit | — |
| Q3_K_Mquality drops | 751.3 GB | ≈ 789.8 | ✕Doesn’t fit | — |
| IQ3_XXSquality drops | 588.0 GB | ≈ 618.3 | ✕Doesn’t fit | — |
| Q2_Kquality drops | 576.4 GB | ≈ 606.2 | ✕Doesn’t fit | — |
| IQ2_XXSquality drops | 395.8 GB | ≈ 416.6 | ✕Doesn’t fit | — |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): no single listed machine; about 14 × 80 GB GPUs.
Run commands
The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 155.9
#37 of 118 listed models · CI 154–157.9
- Epoch AI · Chess Puzzles 46%
#26 of 85 listed models · max effort · ±5.01
- Epoch AI · FrontierMath Tier 4 v2 26.8%
#37 of 65 listed models · max effort · ±7.01
- Epoch AI · FrontierMath Tiers 1–3 64.6%
#37 of 84 listed models · max effort · ±2.84
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 28 – Nov 5, 2026
- LiveBench (White et al.), release 2026-06-25, livebench.ai, licensed CC BY-SA 4.0. This board, adapted from it, is shared under CC BY-SA 4.0. · License · as of Jun 25, 2026
- changes we made
Changes
All changes →- Now priced on Cloudflare Workers AI Input $1.32 · Cache read $0.044 · Output $3.96
- New language model
Sources
- DeepSeek API models & pricing (opens in a new tab) api-docs.deepseek.com · checked Nov 2, 2026
- deepseek-ai/DeepSeek-V4-Pro-0813 model card (opens in a new tab) huggingface.co · checked Nov 2, 2026
- deepseek-ai/DeepSeek-V4-Pro-0813 model card (opens in a new tab) huggingface.co · checked Nov 6, 2026
- Alibaba Cloud Model Studio pricing (opens in a new tab) alibabacloud.com · checked Nov 2, 2026
- 阿里云百炼：模型价格in Chinese (opens in a new tab) help.aliyun.com · checked Nov 2, 2026
- 模型上下架与更新-大模型服务平台百炼-阿里云_大模型服务平台百炼(Model Studio)-大模型服务平台百炼(Model Studio)-阿里云帮助中心in Chinese (opens in a new tab) help.aliyun.com · checked Nov 4, 2026
- Explore Qwen3, Wan2.6 & Latest Models on Model Studio - Alibaba Cloud Model Studio - Alibaba Cloud_Alibaba Cloud Model Studio(Model Studio) - Alibaba Cloud Model Studio - Alibaba Cloud Documentation Center (opens in a new tab) alibabacloud.com · checked Nov 2, 2026
- Cloudflare Workers AI pricing (opens in a new tab) developers.cloudflare.com · checked Nov 1, 2026
- deepseek-ai/DeepSeek-V4-Pro-0813 config.json (opens in a new tab) huggingface.co · checked Nov 4, 2026
- deepseek-ai/DeepSeek-V4-Pro-0813 files (opens in a new tab) huggingface.co · checked Nov 1, 2026
- deepseek-ai/DeepSeek-V4-Pro-0813 at main (opens in a new tab) huggingface.co · checked Nov 1, 2026
DeepSeek’s official policies: data use, privacy, terms →
DeepSeek · China · deepseek.com
Related models
Snapshots, the same family and other models from this provider