Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · EpistemeAI

Reasoning-Llama-3.1-CoT-RE1

Reasoning-Llama-3.1-CoT-RE1 is an 8.03B-parameter Llama-3.1 reasoning fine-tune from EpistemeAI. According to the model card, it uses Unsloth and TRL from a 4-bit Llama instruct base.

Publisher
EpistemeAI
Task
text-generation
Model type
llama
License
llama3.1
Library
transformers
Publication status
Accepted · not indexed

Model overview

Reasoning-Llama-3.1-CoT-RE1 is published by EpistemeAI as a Llama-based text-generation fine-tune. The captured configuration identifies LlamaForCausalLM and Safetensors metadata reports 8030261248 parameters. According to the model card, it was finetuned from unsloth/meta-llama-3.1-8b-instruct-bnb-4bit with Unsloth and TRL, referencing Open-R1.

Recorded capabilities

Llama-3.1 reasoning lineage

According to the model card, it was finetuned from unsloth/meta-llama-3.1-8b-instruct-bnb-4bit as an open reasoning Llama referencing Open-R1.

Unsloth and TRL training

According to the model card, training was 2x faster with Unsloth and Hugging Face TRL, with hub tags listing the Bespoke-Stratos-17k dataset.

Llama architecture and size

Captured config reports LlamaForCausalLM and llama, with Safetensors metadata reporting about 8.03B parameters.

Use cases in the source record

  • Reasoning-style text-generation experiments following the publisher Open-R1 reproduction framing with Unsloth and TRL tooling.

Limitations and unknowns

  • Recorded license is inconsistent: card data says llama3.1 while the card text says apache-2.0, so file-level confirmation is needed.
  • No evaluation results were extracted from this record.
  • No context-window value was extracted from this record.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: EpistemeAI/Reasoning-Llama-3.1-CoT-RE1

Captured: Unknown. Processed: 2026-09-07T19:35:05.149934+00:00.

Reasoning Llama model Open Reasoning Llama model Reference Open-R1: a fully open reproduction of DeepSeek-R1 Uploaded model Developed by: EpistemeAI License: apache-2.0 Finetuned from model : unsloth/meta-llama-3.1-8b-instruct-bnb-4bit This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

F001F002F003F004F005F006F007F009F010F011