Fastest & Best quality & Largest context
gpt-oss:20b
20.9B
120.7 tok/sQuality 76.8Context 8,192 tokens
Highest token generation speed at 120.7 tok/s. Highest completed quality score at 76.8. Largest context that fits in 88GB VRAM, at 8,192 tokens.
Hardware Benchmarks
Best Local LLM Performance for 2× L40s
Compare 2× L40s benchmark results and see which models actually earn the best completed quality scores.
Hardware Overview
GPU-specific specifications for local LLM planning.
- VRAM
- 88GB
- Architecture
- Multi-GPU
Recommended Models
Models we recommend for 2× L40s.
Benchmark Results
Completed quality-ranked benchmark results for 2× NVIDIA L40S.