DeepSeek V4 Pro 0813
Q4 GGUF download: unsloth/DeepSeek-V4-Pro-0813-GGUF - 0.01 GB on disk, read 2026-09-27. VRAM figures above are estimates; this is the actual file size.
Intelligence benchmarks
Artificial Analysis indexes - compared with the best open and proprietary models
Intelligence
36.0
AA Index
Coding
68.8
AA Index
Agentic
41.3
AA Index
Intelligence Index - DeepSeek V4 Pro 0813 vs. the field
Best open-weight models (you can run locally) and leading proprietary models for context.
Coding Index comparison
Agentic Index comparison
Benchmark data from Artificial Analysis · updated 2026-10-01.
Hosted API cost & speed
What it costs to rent this model instead of running it locally
| Metric | Value |
|---|---|
| Input (per 1M tokens) | $1.32 |
| Output (per 1M tokens) | $3.96 |
| Median speed (hosted API) | 81 tok/s |
Median across hosted providers, measured by Artificial Analysis - updated 2026-10-01. Local llama.cpp speed depends on your hardware — see the hardware pages for estimates.
Standard benchmarks
Performance across standard evaluations
| Benchmark | Score |
|---|---|
| GPQA Diamond | 92.8 |
| HLE | 41 |
| SciCode | 51 |
Source: Artificial Analysis component evals - updated 2026-10-01.
Will it run on your hardware?
Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest
Need an exact figure for your context length? Use the VRAM calculator.
Run it locally
Copy-paste - running in under a minute
vllm serve deepseek-ai/DeepSeek-V4-Pro-0813New to this? Start with Ollama · serve to many users with vLLM.
Deep dive
Notes, sources, and the full write-up
DeepSeek V4 Pro 0813 is a 1650.5B-parameter mit model from DeepSeek. It scores 36 on the Artificial Analysis Intelligence Index. At Q4_K_M it needs roughly 957 GB of VRAM, placing it in the 48 GB+ / multi-GPU hardware tier.
Benchmarks
Artificial Analysis Intelligence Index - DeepSeek V4 Pro 0813 vs. leading closed models:
| Model | Intelligence | Coding | GPQA |
|---|---|---|---|
| DeepSeek V4 Pro 0813 | 53.2 | - | 92.8 |
| Claude Opus 5 (max) | 63.1 | - | 93.2 |
| Claude Fable 5 (with fallback) | 62.1 | - | 92.6 |
| GPT-5.6 Sol (max) | 60.9 | - | 94.1 |
| Grok 4.6 (high) | 60.9 | - | 94.9 |
| Qwen3.8 Max | 58.1 | - | 92.7 |
Source: Artificial Analysis (2026-08-15).
Popularity
DeepSeek V4 Pro 0813 has 245 downloads in the last month on HuggingFace and 433 likes.
Frequently asked
Quick answers to common questions
How much VRAM does DeepSeek V4 Pro 0813 need?
DeepSeek V4 Pro 0813 with 1650.5B parameters needs approximately 957 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.
Is DeepSeek V4 Pro 0813 better than other deepseek-ai models?
DeepSeek V4 Pro 0813 has 1650.5B parameters with 1,048,576 context - a strong choice for general use.
What license is DeepSeek V4 Pro 0813 under?
DeepSeek V4 Pro 0813 is released under the mit license, making it suitable for most commercial and personal projects.
What hardware runs DeepSeek V4 Pro 0813 well?
With 1650.5B parameters, DeepSeek V4 Pro 0813 requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.
What is the best quantization for DeepSeek V4 Pro 0813?
Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~1172 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.
How long can DeepSeek V4 Pro 0813's context window handle?
DeepSeek V4 Pro 0813 supports a 1,048,576-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.
What models compete with DeepSeek V4 Pro 0813?
DeepSeek V4 Pro 0813 competes with other 825B–2476B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.
Nearby options
Similar models and compatible hardware by spec