Phind CodeLlama 34B — 34B Dense.
Run Phind CodeLlama 34B locally with Ollama — needs 21.3 GB VRAM at Q4_K_M:
ollama run phind-codellama:34b| Quant | Bits | VRAM @ 16K | Quality |
|---|---|---|---|
| IQ3_XXS | 3.25 | 16.6 GB 14.3 + 2.3 KV | low |
| IQ3_XS | 3.5 | 17.6 GB 15.4 + 2.3 KV | low |
| Q3_K_S | 3.64 | 18.2 GB 16.0 + 2.3 KV | low |
| IQ3_M | 3.76 | 18.7 GB 16.5 + 2.3 KV | low |
| Q3_K_M | 4 | 19.7 GB 17.5 + 2.3 KV | low |
| Q3_K_L | 4.3 | 21.0 GB 18.8 + 2.3 KV | moderate |
| IQ4_XS | 4.46 | 21.7 GB 19.4 + 2.3 KV | moderate |
| Q4_K_S | 4.67 | 22.6 GB 20.3 + 2.3 KV | moderate |
| Q4_K_M | 4.89 | 23.5 GB 21.3 + 2.3 KV | good |
| Q5_K_S | 5.57 | 26.4 GB 24.2 + 2.3 KV | good |
| Q5_K_M | 5.7 | 27.0 GB 24.7 + 2.3 KV | good |
| Q6_K | 6.56 | 30.6 GB 28.4 + 2.3 KV | excellent |
| Q8_0 | 8.5 | 38.9 GB 36.6 + 2.3 KV | lossless |
| FP16 | 16 | 70.7 GB 68.5 + 2.3 KV | lossless |
Select your GPU above to see speed estimates and compatibility for each quantization.
Rent a GPU and run Phind CodeLlama 34B now
Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.
Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.
At Q4_K_M quantization. Sorted by minimum VRAM.
Find the best GPU for Phind CodeLlama 34B
Build Hardware for Phind CodeLlama 34BPhind CodeLlama 34B — fine-tuned for real-world programming tasks and debugging.
Run Phind CodeLlama 34B locally with Ollama — needs 21.3 GB VRAM at Q4_K_M:
ollama run phind-codellama:34b