FITMYLLM · AUGUST 17, 2026 · OPEN METHODOLOGY · COMMUNITY BENCHMARKS
LIVE · UPDATED DAILY
389B
Parameters (52B active)
0
Quantization Options Select your GPU for speed estimates Context length: 4K 8K 16K 32K 64K 128K
Quant Bits VRAM @ 16K Quality IQ2_XXS 2.38 119.2 GB
116.2 + 3.0 KV
low IQ2_M 2.93 146.0 GB
143.0 + 3.0 KV
low Q2_K 3.16 157.1 GB
154.1 + 3.0 KV
low IQ3_XXS 3.25 161.5 GB
158.5 + 3.0 KV
low IQ3_XS 3.5 173.7 GB
170.7 + 3.0 KV
low Q3_K_S 3.64 180.5 GB
177.5 + 3.0 KV
low IQ3_M 3.76 186.3 GB
183.3 + 3.0 KV
low Q3_K_M 4 198.0 GB
195.0 + 3.0 KV
low Q3_K_L 4.3 212.6 GB
209.6 + 3.0 KV
moderate IQ4_XS 4.46 220.4 GB
217.4 + 3.0 KV
moderate Q4_K_S 4.67 230.6 GB
227.6 + 3.0 KV
moderate Q4_K_M 4.89 241.3 GB
238.3 + 3.0 KV
good Q5_K_S 5.57 274.3 GB
271.3 + 3.0 KV
good Q5_K_M 5.7 280.7 GB
277.7 + 3.0 KV
good Q6_K 6.56 322.5 GB
319.5 + 3.0 KV
excellent Q8_0 8.5 416.8 GB
413.8 + 3.0 KV
lossless FP16 16 781.5 GB
778.5 + 3.0 KV
lossless
Select your GPU above to see speed estimates and compatibility for each quantization.
Too big for a single GPU — plan a multi-GPU deployment
Even the lightest quant needs ~119 GB. Size GPUs, replicas, TCO and scaling for a production setup. Open in Enterprise →
▸ READY TO RUN THIS? RENT BY THE HOUR
RENT A GPU AND RUN HUNYUAN-LARGE NOW
Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.
Community Ratings Chat Coding Reasoning Creative Vision Roleplay Agentic
Loading ratings...
Run this model IQ2_XXS — 116.2 GB VRAM IQ2_M — 143.0 GB VRAM Q2_K — 154.1 GB VRAM IQ3_XXS — 158.5 GB VRAM IQ3_XS — 170.7 GB VRAM Q3_K_S — 177.5 GB VRAM IQ3_M — 183.3 GB VRAM Q3_K_M — 195.0 GB VRAM Q3_K_L — 209.6 GB VRAM IQ4_XS — 217.4 GB VRAM Q4_K_S — 227.6 GB VRAM Q4_K_M — 238.3 GB VRAM Q5_K_S — 271.3 GB VRAM Q5_K_M — 277.7 GB VRAM Q6_K — 319.5 GB VRAM Q8_0 — 413.8 GB VRAM FP16 — 778.5 GB VRAM
Ollama llama.cpp vLLM LM Studio KoboldCpp Jan Docker
▸ Easiest way to get started · Beginners
DOCS ↗ curl -fsSL https://ollama.com/install.sh | shCOPY
$ ollama run hunyuan:389b-q4_K_MCOPY
Tag may need adjustment — check ollama.com/library/hunyuan 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.
Auto-detect GPU Live tok/s in chat Speed benchmarks 9 inference engines
GPUs that can run this model At Q4_K_M quantization. Sorted by minimum VRAM.
Find the best GPU for Hunyuan-Large
Build Hardware for Hunyuan-Large ▸ COLOPHON FITMYLLM · INDEPENDENT · DATA-DRIVEN
FITMYLLM · EST. 2025 · © 2026
RECOMMENDATIONS FROM PUBLISHED MATH, CORRECTED BY THE COMMUNITY — 30.
▸ SPEC SHEET
Hunyuan-Large — 389B MoE. ▸ SPECIFICATIONS
PARAMETERS 389B (52B active)
ARCHITECTURE Mixture of Experts
CONTEXT LENGTH 128K tokens
CAPABILITIES chat, coding, reasoning, multilingual, math
RELEASE DATE 2024-11-05
PROVIDER Tencent
FAMILY hunyuan ▸ VRAM REQUIREMENTS
QUANT BPW VRAM QUALITY IQ2_XXS 2.38 116.2 GB 65% IQ2_M 2.93 143.0 GB 75% Q2_K 3.16 154.1 GB 78% IQ3_XXS 3.25 158.5 GB 82% IQ3_XS 3.5 170.7 GB 84% Q3_K_S 3.64 177.5 GB 85% IQ3_M 3.76 183.3 GB 86% Q3_K_M 4 195.0 GB 88% Q3_K_L 4.3 209.6 GB 90% IQ4_XS 4.46 217.4 GB 92% Q4_K_S 4.67 227.6 GB 93% Q4_K_M 4.89 238.3 GB 94% Q5_K_S 5.57 271.3 GB 96% Q5_K_M 5.7 277.7 GB 96% Q6_K 6.56 319.5 GB 97% Q8_0 8.5 413.8 GB 100% FP16 16 778.5 GB 100%
§ 01 BENCHMARK SCORES
HumanEval 90.0
MMLU-PRO 60.2
MATH 77.4
IFEval 85.0
BBH 89.5
MBPP 72.6
GPQA Diamond 42.4
§ 03 COMPATIBLE GPUs
5 @ Q4_K_M Feedback