Zack Zhiyuan Li
commited on
Commit
•
c103f91
1
Parent(s):
264e91e
add README
Browse files
README.md
CHANGED
|
@@ -1,3 +1,70 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
model_name: Octopus-V2-2B
|
| 6 |
+
base_model: NexaAIDev/Octopus-v4
|
| 7 |
+
inference: false
|
| 8 |
+
model_creator: NexaAIDev
|
| 9 |
+
quantized_by: Second State Inc.
|
| 10 |
+
tags:
|
| 11 |
+
- function calling
|
| 12 |
+
- on-device language model
|
| 13 |
+
- android
|
| 14 |
---
|
| 15 |
+
|
| 16 |
+
# Octopus-v4-GGUF
|
| 17 |
+
|
| 18 |
+
## Original Model
|
| 19 |
+
|
| 20 |
+
[NexaAIDev/Octopus-v4](https://huggingface.co/NexaAIDev/Octopus-v4)
|
| 21 |
+
|
| 22 |
+
## Run with [Ollama](https://github.com/ollama/ollama)
|
| 23 |
+
|
| 24 |
+
```bash
|
| 25 |
+
ollama run NexaAIDev/octopus-v4-q4_k_m
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
Input example:
|
| 29 |
+
|
| 30 |
+
```json
|
| 31 |
+
Query: Tell me the result of derivative of x^3 when x is 2?
|
| 32 |
+
|
| 33 |
+
# <nexa_4> represents the math gpt.
|
| 34 |
+
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>
|
| 35 |
+
|
| 36 |
+
```
|
| 37 |
+
|
| 38 |
+
### Dataset and Benchmark
|
| 39 |
+
|
| 40 |
+
* Utilized questions from [MMLU](https://github.com/hendrycks/test) to evaluate the performances.
|
| 41 |
+
* Evaluated with the Ollama [llm-benchmark](https://github.com/MinhNgyuen/llm-benchmark) method.
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
## Quantized GGUF Models
|
| 45 |
+
|
| 46 |
+
| Name | Quant method | Bits | Size | Respons (token/second) | Use Cases |
|
| 47 |
+
| ---------------------- | ------------ | ---- | ------- | ---------------------- | ----------------------------------------- |
|
| 48 |
+
| Octopus-v4.gguf | | | 7.20 GB | 27.64 | extremely large |
|
| 49 |
+
| Octopus-v4-Q2_K.gguf | Q2_K | 2 | 1.32 GB | 54.20 | extremely not recommended, high loss |
|
| 50 |
+
| Octopus-v4-Q3_K.gguf | Q3_K | 3 | 1.82 GB | 51.22 | not recommended |
|
| 51 |
+
| Octopus-v4-Q3_K_S.gguf | Q3_K_S | 3 | 1.57 GB | 51.78 | not very recommended |
|
| 52 |
+
| Octopus-v4-Q3_K_M.gguf | Q3_K_M | 3 | 1.82 GB | 50.86 | not very recommended |
|
| 53 |
+
| Octopus-v4-Q3_K_L.gguf | Q3_K_L | 3 | 1.94 GB | 50.05 | not very recommended |
|
| 54 |
+
| Octopus-v4-Q4_0.gguf | Q4_0 | 4 | 2.03 GB | 65.76 | good quality, recommended |
|
| 55 |
+
| Octopus-v4-Q4_1.gguf | Q4_1 | 4 | 2.24 GB | 69.01 | slow, good quality, recommended |
|
| 56 |
+
| Octopus-v4-Q4_K.gguf | Q4_K | 4 | 2.23 GB | 55.76 | slow, good quality, recommended |
|
| 57 |
+
| Octopus-v4-Q4_K_S.gguf | Q4_K_S | 4 | 2.04 GB | 53.98 | high quality, recommended |
|
| 58 |
+
| Octopus-v4-Q4_K_M.gguf | Q4_K_M | 4 | 1.51 GB | 58.39 | some functions loss, not very recommended |
|
| 59 |
+
| Octopus-v4-Q5_0.gguf | Q5_0 | 5 | 2.45 GB | 61.98 | slow, good quality |
|
| 60 |
+
| Octopus-v4-Q5_1.gguf | Q5_1 | 5 | 2.67 GB | 63.44 | slow, good quality |
|
| 61 |
+
| Octopus-v4-Q5_K.gguf | Q5_K | 5 | 2.58 GB | 58.28 | moderate speed, recommended |
|
| 62 |
+
| Octopus-v4-Q5_K_S.gguf | Q5_K_S | 5 | 2.45 GB | 59.95 | moderate speed, recommended |
|
| 63 |
+
| Octopus-v4-Q5_K_M.gguf | Q5_K_M | 5 | 2.62 GB | 53.31 | fast, good quality, recommended |
|
| 64 |
+
| Octopus-v4-Q6_K.gguf | Q6_K | 6 | 2.91 GB | 52.15 | large, not very recommended |
|
| 65 |
+
| Octopus-v4-Q8_0.gguf | Q8_0 | 8 | 3.78 GB | 50.10 | very large, good quality |
|
| 66 |
+
| Octopus-v4-f16.gguf | f16 | 16 | 7.20 GB | 30.61 | extremely large |
|
| 67 |
+
|
| 68 |
+
_Quantized with llama.cpp_
|
| 69 |
+
|
| 70 |
+
|