Will It Run AI
CalculatorModelsHardwareCompare
Product
  • Calculator
  • Compare
  • Tier List
Browse
  • Models
  • Hardware
  • Docs
About
  • Why It Works
  • What's New
  • Legal Notice
  • Privacy Policy

All estimates are approximations based on mathematical models and public specifications. Actual performance may vary. Do not make purchasing decisions based solely on these estimates.

Data sourced from Hugging Face, Ollama, and official model documentation. Model names and logos are trademarks of their respective owners.

© 2026 Will It Run AI — Fase Consulting Ibiza, S.L. (NIF: B57969656)

Home/Meta Llama 3 8B Instruct/on NVIDIA H20 96GB

Can it run?

Can NVIDIA H20 96GB run Meta Llama 3 8B Instruct?

CUsable

Runs well

Using Q4_K_M in Ollama

Capabilities:

Fit status

Runs well

Decode

663.9 tok/s

TTFT

350 ms

Safe context

91K

Memory

16.9 GB / 96.0 GB

Memory breakdown

Weights4.9 GB
KV Cache1.3 GB
Runtime1.2 GB
Headroom9.6 GB

Performance by workload

WorkloadGradeFitDecodeTTFTContext
Agentic CodingCRuns well663.9 tok/s424 ms169K
ChatCRuns well663.9 tok/s350 ms47K
CodingCRuns well663.9 tok/s350 ms91K
RAGCRuns well663.9 tok/s530 ms169K
ReasoningCRuns well663.9 tok/s350 ms91K

Quantization options

How Meta Llama 3 8B Instruct (8B params) fits at each quantization level on NVIDIA H20 96GB (96.0 GB usable).

QuantBitsVRAMQualityFit
Q2_K
2
3.1 GB
LowD30
Q3_K_S
3
3.9 GB
LowD30
NVFP4
4
4.5 GB
MediumD30
Q4_K_M
4
4.9 GB
MediumD30
Q5_K_M
5
5.8 GB
HighD31
Q6_K
6
6.6 GB
HighD31
Q8_0
8
8.6 GB
Very HighD31
F16Best for your GPU
16
16.4 GB
MaximumD33

Get started

Upgrade options

Hardware that runs Meta Llama 3 8B Instruct well

AppleMac Studio M3 Ultra 256GBBudget pick
C114.1 tok/s decode

 

See all results for NVIDIA H20 96GBSee all hardware for Meta Llama 3 8B Instruct