reader-lm-1.5b-IMat-GGUF

Llama.cpp imatrix quantization of jinaai/reader-lm-1.5b

Original Model: jinaai/reader-lm-1.5b
Original dtype: BF16 (bfloat16)
Quantized by: llama.cpp b3731
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
reader-lm-1.5b.Q8_0.gguf Q8_0 1.65GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q6_K.gguf Q6_K 1.27GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q4_K.gguf Q4_K 986.05MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q3_K.gguf Q3_K 824.18MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q2_K.gguf Q2_K 676.30MB βœ… Available 🟒 IMatrix πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
reader-lm-1.5b.Q8_0.gguf Q8_0 1.65GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q6_K.gguf Q6_K 1.27GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q5_K.gguf Q5_K 1.13GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q5_K_S.gguf Q5_K_S 1.10GB βœ… Available βšͺ Static πŸ“¦ No
reader-lm-1.5b.Q4_K.gguf Q4_K 986.05MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q4_K_S.gguf Q4_K_S 940.31MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ4_NL.gguf IQ4_NL 936.33MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ4_XS.gguf IQ4_XS 895.73MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q3_K.gguf Q3_K 824.18MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q3_K_L.gguf Q3_K_L 880.16MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q3_K_S.gguf Q3_K_S 760.94MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ3_M.gguf IQ3_M 776.66MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ3_S.gguf IQ3_S 762.40MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ3_XS.gguf IQ3_XS 731.70MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ3_XXS.gguf IQ3_XXS 668.79MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q2_K.gguf Q2_K 676.30MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.Q2_K_S.gguf Q2_K_S 640.13MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ2_M.gguf IQ2_M 601.05MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ2_S.gguf IQ2_S 563.81MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ2_XS.gguf IQ2_XS 550.32MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ2_XXS.gguf IQ2_XXS 511.02MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ1_M.gguf IQ1_M 464.46MB βœ… Available 🟒 IMatrix πŸ“¦ No
reader-lm-1.5b.IQ1_S.gguf IQ1_S 436.53MB βœ… Available 🟒 IMatrix πŸ“¦ No

Downloading using huggingface-cli

If you do not have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Download the specific file you want:

huggingface-cli download legraphista/reader-lm-1.5b-IMat-GGUF --include "reader-lm-1.5b.Q8_0.gguf" --local-dir ./

If the model file is big, it has been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download legraphista/reader-lm-1.5b-IMat-GGUF --include "reader-lm-1.5b.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Simple chat template

<|im_start|>system
You are a helpful assistant.<|im_end|>
<|im_start|>user
{user_prompt}<|im_end|>
<|im_start|>assistant
{assistant_response}<|im_end|>
<|im_start|>user
{next_user_prompt}<|im_end|>

Chat template with system prompt

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{user_prompt}<|im_end|>
<|im_start|>assistant
{assistant_response}<|im_end|>
<|im_start|>user
{next_user_prompt}<|im_end|>

Llama.cpp

llama.cpp/main -m reader-lm-1.5b.Q8_0.gguf --color -i -p "prompt here (according to the chat template)"

FAQ

Why is the IMatrix not applied everywhere?

According to this investigation, it appears that lower quantizations are the only ones that benefit from the imatrix input (as per hellaswag results).

How do I merge a split GGUF?

  1. Make sure you have gguf-split available
  2. Locate your GGUF chunks folder (ex: reader-lm-1.5b.Q8_0)
  3. Run gguf-split --merge reader-lm-1.5b.Q8_0/reader-lm-1.5b.Q8_0-00001-of-XXXXX.gguf reader-lm-1.5b.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!

Downloads last month
14
GGUF
Model size
1.54B params
Architecture
qwen2

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Examples
Inference API (serverless) has been turned off for this model.

Model tree for legraphista/reader-lm-1.5b-IMat-GGUF

Quantized
(10)
this model