FitMyLLM
DeepSeek/Mixture of Experts

DeepSeekDeepSeek V4 Pro

chatcodingreasoningmultilingualmathagentictool_use
1600B
Parameters (49B active)
1024K
Context length
8
Benchmarks
17
Quantizations
174K
HF downloads
Architecture
MoE
Released
2026-04-24
Layers
61
KV Heads
1
Head Dim
512
Family
deepseek

Quantization Options

QuantBitsVRAMQuality
IQ2_XXS2.38476.5 GBlow
IQ2_M2.93586.5 GBlow
Q2_K3.16632.5 GBlow
IQ3_XXS3.25650.5 GBlow
IQ3_XS3.5700.5 GBlow
Q3_K_S3.64728.5 GBlow
IQ3_M3.76752.5 GBlow
Q3_K_M4800.5 GBlow
Q3_K_L4.3860.5 GBmoderate
IQ4_XS4.46892.5 GBmoderate
Q4_K_S4.67934.5 GBmoderate
Q4_K_M4.89978.5 GBgood
Q5_K_S5.571114.5 GBgood
Q5_K_M5.71140.5 GBgood
Q6_K6.561312.5 GBexcellent
Q8_08.51700.5 GBlossless
FP16163200.5 GBlossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN DEEPSEEK V4 PRO NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (8)

LiveCodeBench93.5
GPQA Diamond90.1
MMLU-PRO87.5
BBH87.5
SWE-bench80.6
HumanEval76.8
Terminal-Bench67.9
HLE37.7

Run this model

▸Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run deepseek:1600b-q4_K_M

Tag may need adjustment — check ollama.com/library/deepseek for available tags.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines
HuggingFaceGGUF DownloadsBuild Hardware

Find the best GPU for DeepSeek V4 Pro

Build Hardware for DeepSeek V4 Pro
▸ SPEC SHEET

DeepSeek V4 Pro — 1600B MoE.

▸ SPECIFICATIONS
PARAMETERS
1600B (49B active)
ARCHITECTURE
Mixture of Experts
CONTEXT LENGTH
1024K tokens
CAPABILITIES
chat, coding, reasoning, multilingual, math, agentic, tool_use
RELEASE DATE
2026-04-24
PROVIDER
DeepSeek
FAMILY
deepseek
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ2_XXS2.38476.5 GB65%
IQ2_M2.93586.5 GB75%
Q2_K3.16632.5 GB78%
IQ3_XXS3.25650.5 GB82%
IQ3_XS3.5700.5 GB84%
Q3_K_S3.64728.5 GB85%
IQ3_M3.76752.5 GB86%
Q3_K_M4800.5 GB88%
Q3_K_L4.3860.5 GB90%
IQ4_XS4.46892.5 GB92%
Q4_K_S4.67934.5 GB93%
Q4_K_M4.89978.5 GB94%
Q5_K_S5.571114.5 GB96%
Q5_K_M5.71140.5 GB96%
Q6_K6.561312.5 GB97%
Q8_08.51700.5 GB100%
FP16163200.5 GB100%
§ 01BENCHMARK SCORES
HumanEval76.8
MMLU-PRO87.5
BBH87.5
GPQA Diamond90.1
LiveCodeBench93.5
HLE37.7
aa_terminal_bench67.9
SWE-bench80.6