Run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF locally

⬇ 163,018 ❤ 18
Parameters1.48B
Context131,072

MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF is a compact code-focused language model with 1.48 billion parameters, built on the llama architecture. It has been downloaded 163,018 times.

To run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF locally at a 4,096-token context, its quantized versions need between 2.01 GB (IQ1_S, lowest quality) and 4.3 GB (GGUF, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is Q5_K_M, needing about 2.57 GB. That means MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

All quantizations

Quant.Bits QualityWeights KVTotal Speed~Verdict
IQ1_S 2.66 Low 0.46 GB 0.75 GB 2.01 GB 874.4 t/s Fits in VRAM
IQ1_M 2.76 Low 0.47 GB 0.75 GB 2.02 GB 844.5 t/s Fits in VRAM
IQ2_XS 3.06 Low 0.53 GB 0.75 GB 2.08 GB 761.6 t/s Fits in VRAM
Q2_K 3.44 Fair 0.59 GB 0.75 GB 2.14 GB 676.7 t/s Fits in VRAM
IQ3_XS 3.77 Fair 0.65 GB 0.75 GB 2.2 GB 618.0 t/s Fits in VRAM
Q3_K_S 3.92 Fair 0.67 GB 0.75 GB 2.22 GB 593.7 t/s Fits in VRAM
Q3_K_M 4.26 Good 0.73 GB 0.75 GB 2.28 GB 546.6 t/s Fits in VRAM
Q3_K_L 4.48 Good 0.77 GB 0.75 GB 2.32 GB 519.9 t/s Fits in VRAM
IQ4_XS 4.51 Good 0.78 GB 0.75 GB 2.33 GB 515.9 t/s Fits in VRAM
Q4_K_S 4.9 Good 0.84 GB 0.75 GB 2.39 GB 475.0 t/s Fits in VRAM
Q4_K_M 5.22 Very good 0.9 GB 0.75 GB 2.45 GB 445.7 t/s Fits in VRAM
Q5_K_S 5.7 Very good 0.98 GB 0.75 GB 2.53 GB 408.6 t/s Fits in VRAM
Q5_K_M 5.96 Very good 1.02 GB 0.75 GB 2.57 GB 390.4 t/s Fits in VRAM
GGUF 16.01 Excellent 2.75 GB 0.75 GB 4.3 GB 18.2 t/s Offload

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF?

You need about 4.3 GB of VRAM to run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF entirely on the GPU using the GGUF quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF fully on the GPU using GGUF (about 4.3 GB).

Can I run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF fully on the GPU using GGUF (about 4.3 GB).

Can I run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF fully on the GPU using GGUF (about 4.3 GB).

What is the best quantization for MaziyarPanahi/Yi-Coder-1.5B-Chat-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.