legraphista's picture
Upload README.md with huggingface_hub
4ef5488 verified
metadata
base_model: Salesforce/xLAM-8x7b-r
datasets:
  - Salesforce/xlam-function-calling-60k
extra_gated_button_content: Agree and access repository
extra_gated_heading: Acknowledge to follow corresponding license to access the repository
inference: false
language:
  - en
library_name: gguf
license: cc-by-nc-4.0
pipeline_tag: text-generation
quantized_by: legraphista
tags:
  - function-calling
  - LLM Agent
  - tool-use
  - mistral
  - pytorch
  - quantized
  - GGUF
  - quantization
  - imat
  - imatrix
  - static
  - 16bit
  - 8bit
  - 6bit
  - 5bit
  - 4bit
  - 3bit
  - 2bit
  - 1bit

xLAM-8x7b-r-IMat-GGUF

Llama.cpp imatrix quantization of Salesforce/xLAM-8x7b-r

Original Model: Salesforce/xLAM-8x7b-r
Original dtype: BF16 (bfloat16)
Quantized by: llama.cpp b3647
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
xLAM-8x7b-r.Q8_0/* Q8_0 49.63GB βœ… Available βšͺ Static βœ‚ Yes
xLAM-8x7b-r.Q6_K.gguf Q6_K 38.38GB βœ… Available βšͺ Static πŸ“¦ No
xLAM-8x7b-r.Q4_K.gguf Q4_K 28.45GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q3_K.gguf Q3_K 22.55GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q2_K.gguf Q2_K 17.31GB βœ… Available 🟒 IMatrix πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
xLAM-8x7b-r.BF16/* BF16 93.41GB βœ… Available βšͺ Static βœ‚ Yes
xLAM-8x7b-r.FP16/* F16 93.41GB βœ… Available βšͺ Static βœ‚ Yes
xLAM-8x7b-r.Q8_0/* Q8_0 49.63GB βœ… Available βšͺ Static βœ‚ Yes
xLAM-8x7b-r.Q6_K.gguf Q6_K 38.38GB βœ… Available βšͺ Static πŸ“¦ No
xLAM-8x7b-r.Q5_K.gguf Q5_K 33.23GB βœ… Available βšͺ Static πŸ“¦ No
xLAM-8x7b-r.Q5_K_S.gguf Q5_K_S 32.23GB βœ… Available βšͺ Static πŸ“¦ No
xLAM-8x7b-r.Q4_K.gguf Q4_K 28.45GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q4_K_S.gguf Q4_K_S 26.75GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ4_NL.gguf IQ4_NL 26.51GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ4_XS.gguf IQ4_XS 25.08GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q3_K.gguf Q3_K 22.55GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q3_K_L.gguf Q3_K_L 24.17GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q3_K_S.gguf Q3_K_S 20.43GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ3_M.gguf IQ3_M 21.43GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ3_S.gguf IQ3_S 20.43GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ3_XS.gguf IQ3_XS 19.35GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ3_XXS.gguf IQ3_XXS 18.24GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q2_K.gguf Q2_K 17.31GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.Q2_K_S.gguf Q2_K_S 16.03GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ2_M.gguf IQ2_M 15.50GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ2_S.gguf IQ2_S 14.13GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ2_XS.gguf IQ2_XS 13.92GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ2_XXS.gguf IQ2_XXS 12.56GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ1_M.gguf IQ1_M 10.85GB βœ… Available 🟒 IMatrix πŸ“¦ No
xLAM-8x7b-r.IQ1_S.gguf IQ1_S 9.82GB 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/xLAM-8x7b-r-IMat-GGUF --include "xLAM-8x7b-r.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/xLAM-8x7b-r-IMat-GGUF --include "xLAM-8x7b-r.Q8_0/*" --local-dir ./
# see FAQ for merging GGUF's

Inference

Simple chat template

<s> [INST] {user_prompt} [/INST] {assistant_response}</s> [INST] {next_user_prompt} [/INST]

Chat template with system prompt

<s> [INST] {system_prompt}

{user_prompt} [/INST] {assistant_response}</s> [INST] {next_user_prompt} [/INST]

Llama.cpp

llama.cpp/main -m xLAM-8x7b-r.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: xLAM-8x7b-r.Q8_0)
  3. Run gguf-split --merge xLAM-8x7b-r.Q8_0/xLAM-8x7b-r.Q8_0-00001-of-XXXXX.gguf xLAM-8x7b-r.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!