octo-net-gguf / README.md
zackli4ai's picture
Update README.md
e046e74 verified
|
raw
history blame
4.67 kB
metadata
language:
  - en
license: cc-by-nc-4.0
model_name: Octopus-V4-GGUF
base_model: NexaAIDev/Octopus-v4
inference: false
model_creator: NexaAIDev
quantized_by: Nexa AI, Inc.
tags:
  - function calling
  - on-device language model
  - gguf
  - llama cpp

Octopus V4-GGUF: Graph of language models

- Original Model - Nexa AI Website - Octopus-v4 Github - ArXiv - Domain LLM Leaderbaord

nexa-octopus

Acknowledgement:
We sincerely thank our community members, Mingyuan and Zoey, for their extraordinary contributions to this quantization effort. Please explore Octopus-v4 for our original huggingface model.

Run with Ollama

ollama run NexaAIDev/octopus-v4-q4_k_m

Input example:

Query: Tell me the result of derivative of x^3 when x is 2?

Response: <nexa_4> ('Determine the derivative of the function f(x) = x^3 at the point where x equals 2, and interpret the result within the context of rate of change and tangent slope.')<nexa_end>

Note that <nexa_4> represents the math gpt.

Dataset and Benchmark

  • Utilized questions from MMLU to evaluate the performances.
  • Evaluated with the Ollama llm-benchmark method.

Quantized GGUF Models

Name Quant method Bits Size Respons (token/second) Use Cases
Octopus-v4.gguf 7.20 GB 27.64 extremely large
Octopus-v4-Q2_K.gguf Q2_K 2 1.32 GB 54.20 extremely not recommended, high loss
Octopus-v4-Q3_K.gguf Q3_K 3 1.82 GB 51.22 not recommended
Octopus-v4-Q3_K_S.gguf Q3_K_S 3 1.57 GB 51.78 not very recommended
Octopus-v4-Q3_K_M.gguf Q3_K_M 3 1.82 GB 50.86 not very recommended
Octopus-v4-Q3_K_L.gguf Q3_K_L 3 1.94 GB 50.05 not very recommended
Octopus-v4-Q4_0.gguf Q4_0 4 2.03 GB 65.76 good quality, recommended
Octopus-v4-Q4_1.gguf Q4_1 4 2.24 GB 69.01 slow, good quality, recommended
Octopus-v4-Q4_K.gguf Q4_K 4 2.23 GB 55.76 slow, good quality, recommended
Octopus-v4-Q4_K_S.gguf Q4_K_S 4 2.04 GB 53.98 high quality, recommended
Octopus-v4-Q4_K_M.gguf Q4_K_M 4 1.51 GB 58.39 some functions loss, not very recommended
Octopus-v4-Q5_0.gguf Q5_0 5 2.45 GB 61.98 slow, good quality
Octopus-v4-Q5_1.gguf Q5_1 5 2.67 GB 63.44 slow, good quality
Octopus-v4-Q5_K.gguf Q5_K 5 2.58 GB 58.28 moderate speed, recommended
Octopus-v4-Q5_K_S.gguf Q5_K_S 5 2.45 GB 59.95 moderate speed, recommended
Octopus-v4-Q5_K_M.gguf Q5_K_M 5 2.62 GB 53.31 fast, good quality, recommended
Octopus-v4-Q6_K.gguf Q6_K 6 2.91 GB 52.15 large, not very recommended
Octopus-v4-Q8_0.gguf Q8_0 8 3.78 GB 50.10 very large, good quality
Octopus-v4-f16.gguf f16 16 7.20 GB 30.61 extremely large

Quantized with llama.cpp