DeepSeek-V3.2
Q4 GGUF download: unsloth/DeepSeek-V3.2-GGUF - 48.47 GB on disk, read 2026-10-10. 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
21.5
AA Index
Coding
44.2
AA Index
Intelligence Index - DeepSeek-V3.2 vs. the field
Best open-weight models (you can run locally) and leading proprietary models for context.
Coding Index comparison
Benchmark data from Artificial Analysis · updated 2026-10-10.
Hosted API cost & speed
What it costs to rent this model instead of running it locally
| Metric | Value |
|---|---|
| Input (per 1M tokens) | $0.28 |
| Output (per 1M tokens) | $0.42 |
Median across hosted providers, measured by Artificial Analysis - updated 2026-10-10. Local llama.cpp speed depends on your hardware — see the hardware pages for estimates.
Arena rating (LMArena)
Human preference wins from head-to-head chat battles
| Metric | Value |
|---|---|
| Overall arena rating | 1424.5 |
| Rank (overall) | #104 |
| Battle votes | 48,114 |
Overall rating published by LMArena on 2026-10-08 - a pairwise human-preference Elo across live chat battles; higher is better. Fetched 2026-10-10. Open-weights models compete head-to-head with proprietary ones here — no benchmark prompts, just what people pick.
Standard benchmarks
Performance across standard evaluations
| Benchmark | Score |
|---|---|
| MMLU-Pro | 85 |
| GPQA Diamond | 82.4 |
| AIME | 94.2 |
| SWEBENCH | 70 |
| HLE | 40.8 |
| Terminal-Bench | 39.6 |
Additional scores from Hugging Face OpenEvals official leaderboards - updated 2026-10-10.
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-V3.2New to this? Start with Ollama · serve to many users with vLLM.
Deep dive
Notes, sources, and the full write-up
Frequently asked
Quick answers to common questions
How much VRAM does DeepSeek-V3.2 need?
DeepSeek-V3.2 with 685.4B parameters needs approximately 398 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.
Is DeepSeek-V3.2 better than other deepseek-ai models?
DeepSeek-V3.2 has 685.4B parameters with 163,840 context - a strong choice for general use.
What license is DeepSeek-V3.2 under?
DeepSeek-V3.2 is released under the mit license, making it suitable for most commercial and personal projects.
What hardware runs DeepSeek-V3.2 well?
With 685.4B parameters, DeepSeek-V3.2 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-V3.2?
Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~487 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.
How long can DeepSeek-V3.2's context window handle?
DeepSeek-V3.2 supports a 163,840-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-V3.2?
DeepSeek-V3.2 competes with other 343B–1028B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.
Nearby options
Similar models and compatible hardware by spec