Fastest
lfm2.5:8b-64k
8.5B
189.9 tok/s
Highest token generation speed at 189.9 tok/s.
Hardware Benchmarks
Best Local LLM Performance for RTX 2070 Super
Compare RTX 2070 Super benchmark results and see which models actually earn the best completed quality scores.
Hardware Overview
GPU-specific specifications for local LLM planning.
- VRAM
- 8GB
Recommended Models
Models we recommend for RTX 2070 Super.
Best quality
bonsai-2-27b
27B
38.3 tok/sQuality 85.0
Highest completed quality score at 85.0.
Largest context
hf.co/unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
7.46B
79.6 tok/sContext 131,072 tokens
Largest context that fits in 8GB VRAM, at 131,072 tokens.
Models on the RTX 2070 Super
Every model the community ran on this hardware. Up is better answers, right is faster. Time to finish counts every token the model wrote, thinking included, so a model that talks a lot ends up further left.
Best trade-off: quality 60+, and nothing else here is both faster and betterOther models · dot size = number of runs
Medians over 11 fully rated runs of 7 models. Each quantization is its own model.
Frequently asked questions
Which models actually run best on the RTX 2070 Super, by task, from community benchmark data.
- What is the best local model for coding on the RTX 2070 Super?
- On the RTX 2070 Super, ggml-org/gemma-4-E4B-it-GGUF:Q4_0 ranks first for coding in our benchmarks (60.2/100 at ~86.9 tok/s).
- What is the best local model for agentic workflows on the RTX 2070 Super?
- On the RTX 2070 Super, ggml-org/gemma-4-E4B-it-GGUF:Q4_0 ranks first for agentic workflows in our benchmarks (86.0/100 at ~86.9 tok/s).
Benchmark Results
Completed quality-ranked benchmark results for NVIDIA GeForce RTX 2070 SUPER.