L
localai.computer
ModelsGPUsSystemsAI SetupsBuildsOpenClawMethodology

Resources

  • Methodology
  • Submit Benchmark
  • About

Browse

  • AI Models
  • GPUs
  • PC Builds

Guides

  • OpenClaw Guide
  • How-To Guides

Legal

  • Privacy
  • Terms
  • Contact

© 2025 localai.computer. Hardware recommendations for running AI models locally.

ℹ️We earn from qualifying purchases through affiliate links at no extra cost to you. This supports our free content and research.

Can NVIDIA L40S run unsloth/gpt-oss-20b-unsloth-bnb-4bit?

Runs Q448GB VRAM availableRequires 10GB+

NVIDIA L40S meets the minimum VRAM requirement for Q4 inference of unsloth/gpt-oss-20b-unsloth-bnb-4bit. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.

What this means for you

NVIDIA L40S can run unsloth/gpt-oss-20b-unsloth-bnb-4bit with Q4 quantization. At approximately 97 tokens/second, you can expect Good speed - acceptable for interactive use.

You have 38GB headroom, which is sufficient for system overhead and smooth operation.

Quantization breakdown

QuantizationVRAM neededVRAM availableEstimated speedVerdict
Q410GB48GB97.08 tok/s✅ Fits comfortably
Q820GB48GB66.02 tok/s✅ Fits comfortably
FP1641GB48GB33.01 tok/s✅ Fits comfortably

Suitable alternatives

AMD Instinct MI300X
192GB
395.16 tok/s
Price: —
NVIDIA H200 SXM 141GB
141GB
393.34 tok/s
Price: —
AMD Instinct MI300X
192GB
316.57 tok/s
Price: —
AMD Instinct MI250X
128GB
278.69 tok/s
Price: —
NVIDIA H100 SXM5 80GB
80GB
245.73 tok/s
Price: —

More questions

NVIDIA L40S specs & pricingFull guide for unsloth/gpt-oss-20b-unsloth-bnb-4bitunsloth/gpt-oss-20b-unsloth-bnb-4bit speed on NVIDIA L40Sunsloth/gpt-oss-20b-unsloth-bnb-4bit Q4 requirements