---
title: Apriel-R1PV.2-Thinking
canonical_url: "https://www.modelscope.cn/models/ConicCat/Apriel-R1PV.2-Thinking"
md_url: "https://www.modelscope.cn/models/ConicCat/Apriel-R1PV.2-Thinking.md"
repository: ConicCat/Apriel-R1PV.2-Thinking
last_updated: 2026-03-19
model_type:
  - mistral
architectures:
  - MistralForCausalLM
base_model:
  - ServiceNow-AI/Apriel-Nemotron-15b-Thinker
base_model_relation: finetune
parameters: 15.0B
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 12
stars: 0
---

# Apriel-R1PV.2-Thinking

> Apriel-R1PV.2-Thinking - ConicCat 在 ModelScope 开源的模型。Day 2 RP finetune of Apriel 15B, with several iterative improvements from the first version. In particular, coherence at good temperatures (.7) should be much higher.

ConicCat/Apriel-R1PV.2-Thinking 是 ModelScope 魔搭社区上的 15.0B 参数机器学习模型，基于 ServiceNow-AI/Apriel-Nemotron-15b-Thinker 构建，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: ConicCat/Apriel-R1PV.2-Thinking
- **Parameters**: 15.0B
- **Base model**: ServiceNow-AI/Apriel-Nemotron-15b-Thinker
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Downloads**: 12
- **Stars**: 0
- **Last updated**: 2026-03-19

Source: https://www.modelscope.cn/models/ConicCat/Apriel-R1PV.2-Thinking

---

### Apriel R1P V.2

Day 2 RP finetune of Apriel 15B, with several iterative improvements from the first version. In particular, coherence at good temperatures (~.7) should be much higher.

I also fully converted the model to use the Phi 3 format; this comes at the slight tradeoff of the `<|end|>` tag not always tokenizing exactly the same way in a few niche scenarios.

Further attempts were made to fix formatting issues with asterisks on the base model.

NOTE: THIS IS THE THINKING VERSION

Upon further testing, I discovered that while merging back onto the instruct improved thinking mode, this came at the cost of degraded non-thinking outputs.

Use the [non-thinking version](https://huggingface.co/ConicCat/Apriel-R1PV.2-NoThink) instead if you want a standard model.

## Thinking Mode

- To enable thinking place `/think` in the system prompt and prefill `<|think|>\n` for thinking mode.

- Phi esque thinking tags, `<|think|>` and `<|/think|>` have been added to the model.

- Remember to reconfigure Sillytavern to parse the new think tags.

## Settings

The chat template has been converted to a Phi 3 template as the model seemed to respond best to this format.

This model does prefer having character cards placed in user messages, not the system prompt.

## Special Thanks:

Undi95 for portions of their dataset and inspiration.

PJMixers-Dev for their dataset curation and creation efforts.