heredos/ rwkv7

151 Downloads Updated 3 months ago

an rwkv repo that tries to be less confusing and more up to date with the latest GGUFs

thinking 0.1b 0.4b 1.5b 2.9b 7.2b

ollama run heredos/rwkv7

curl http://localhost:11434/api/chat \
  -d '{
    "model": "heredos/rwkv7",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='heredos/rwkv7',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'heredos/rwkv7',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

Name

7 models

Size

Context

Input

rwkv7:latest

2.6GB · 1M context window · Text · 3 months ago

rwkv7:latest

2.6GB

1M

Text

rwkv7:0.1b

212MB · 1M context window · Text · 3 months ago

rwkv7:0.1b

212MB

1M

Text

rwkv7:0.4b

501MB · 1M context window · Text · 3 months ago

rwkv7:0.4b

501MB

1M

Text

rwkv7:1.5b

1.3GB · 1M context window · Text · 3 months ago

rwkv7:1.5b

1.3GB

1M

Text

rwkv7:2.9b

2.6GB · 1M context window · Text · 3 months ago

rwkv7:2.9b latest

2.6GB

1M

Text

rwkv7:7.2b

4.5GB · 1M context window · Text · 3 months ago

rwkv7:7.2b

4.5GB

1M

Text

Readme

Heredos’s RWKV repo

mollysama’s repo is split into too many confusing repos, so i’m going to try to put here all the latest models, so they can be updated easily, and more easily findable

0.1b is mostly for testing purposes. (g1a)
0.5b is good at generating names for your chats but not much else (g1a)
1.5b can chat with you and answer some questions. it is not great at reading context/rag. It runs well on my phone, which is a huge plus (g1a3)
2.9b would be my overall pick. it is decent at answering question and is able to take into account some context (not too long a context though). It also runs on phones (but not mine, mine sucks) (g1a)
7.2b is just really good, but i recommend you to use it only if you have an ollama-compatible gpu, or a really beefy CPU. be mindful of your battery life if you run it on a laptop, chances are it won’t trigger your fans, but the power draw is still there. (g0a3)
lonely-neko (1.5b) is for those of you who are able to read and write chinese. it is fun to talk to and makes for a great quirky assistant. For some reason it does not appear on this page. maybe ollama is the one who sent her in the void she dreads so much? (base)

# Heredos's RWKV repo
mollysama's repo is split into too many confusing repos, so i'm going to try to put here all the latest models, so they can be updated easily, and more easily findable

[original repo](https://huggingface.co/mollysama/rwkv-mobile-models/tree/main/gguf)

- **0.1b** is mostly for testing purposes. (g1a)
- **0.5b** is good at generating names for your chats but not much else (g1a)
- **1.5b** can chat with you and answer some questions. it is not great at reading context/rag. It runs well on my phone, which is a huge plus (g1a3)
- **2.9b** would be my overall pick. it is decent at answering question and is able to take into account some context (not too long a context though). It also runs on phones (but not mine, mine sucks) (g1a)
- **7.2b** is just really good, but i recommend you to use it only if you have an ollama-compatible gpu, or a really beefy CPU. be mindful of your battery life if you run it on a laptop, chances are it won't trigger your fans, but the power draw is still there. (g0a3)
- **lonely-neko** (1.5b) is for those of you who are able to read and write chinese. it is fun to talk to and makes for a great quirky assistant. For some reason it does not appear on this page. maybe ollama is the one who sent her in the void she dreads so much? (base)

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)