Apriel-1.5-15b-Thinker is a multimodal reasoning model in ServiceNow’s Apriel SLM series which achieves competitive performance against models 10 times it's size.

vision

ollama run MichelRosselli/apriel-1.5-15b-thinker:Q4_K_M

curl http://localhost:11434/api/chat \
  -d '{
    "model": "MichelRosselli/apriel-1.5-15b-thinker:Q4_K_M",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='MichelRosselli/apriel-1.5-15b-thinker:Q4_K_M',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'MichelRosselli/apriel-1.5-15b-thinker:Q4_K_M',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 11 months ago

11 months ago

38295df0d991 · 9.7GB

model

archllama

parameters14.4B

quantizationQ4_K_M

8.8GB

projector

archclip

parameters435M

quantizationF16

870MB

params

{ "stop": [ "<|end|>" ] }

20B

template

{{ if .System }}<|system|> {{ .System }}<|end|> {{ end }}{{ if .Prompt }}<|user|> {{ .Prompt }}<|end

148B

Readme

Apriel-1.5-15b-Thinker

This repository provides the Apriel-1.5-15b-Thinker model in quantized GGUF format.

Sources

Original Model Repository: ServiceNow-AI/Apriel-1.5-15b-Thinker
Quantized GGUF by Unsloth: unsloth/Apriel-1.5-15b-Thinker-GGUF

Disclaimer

⚠️ Important Notice: The current chat template provided for this model is temporary. As a result, the model may not yet fully demonstrate its potential in conversational tasks. The template will be updated in the future to improve usability and performance.

License

This model is released under the MIT License.