community pool / model answer
Qwen2.5-Coder-14B-Instruct
code · Qwen/Qwen2.5-Coder-14B-Instruct
35 tok/s · model median · 13 runs in the model record.
| Reference rig | Quantization | Median result | Basis | TTFT (ms) | Peak VRAM | Max context | Engines |
|---|---|---|---|---|---|---|---|
| RTX 2080 8GB | 4-bit | 11 tok/s | reported | - | 9 GB | 2,048 | ollama |
| RTX 3060 12GB | 4-bit | 35 tok/s | reported | - | 9 GB | 2,048 | ollama |
| RTX 3080 12GB | 4-bit | 72 tok/s | measured | - | 9 GB | 2,048 | llama.cpp, ollama |
| RTX 5000 16GB | 5-bit | 24.3 tok/s | reported | - | 14.6 GB | 32,768 | llama.cpp |
| RTX PRO 6000 Blackwell 96GB | 16-bit | 50.3 tok/s | reported | - | - GB | 32,768 | vllm |
| RTX3060+RTX2080-pooled-20GB 20GB | 4-bit | 35 tok/s | measured | - | 9 GB | 2,048 | ollama |
Context tested
The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.