unsloth/Llama-3.1-70B

TEXT GENERATIONPricing:Input $2.88 / Output $2.88Concurrent Unit Cost:4Model Size:70BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Nov 20, 2025License:llama3.1Architecture:Transformer Featherless Exclusive Cold

The unsloth/Llama-3.1-70B model is a 70 billion parameter instruction-tuned large language model developed by Meta, part of the Llama 3.1 collection. Optimized with an improved transformer architecture and trained on over 15 trillion tokens, it excels in multilingual dialogue use cases and code generation. This model supports a 128k context length and is designed for commercial and research applications, outperforming many open-source and closed chat models on common industry benchmarks.

Loading preview...

unsloth/Llama-3.1-70B: An Overview

This model is a 70 billion parameter instruction-tuned variant from Meta's Llama 3.1 family, released on July 23, 2024. It leverages an optimized transformer architecture and has been fine-tuned using supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) for enhanced helpfulness and safety. The model was trained on over 15 trillion tokens of publicly available online data, with a knowledge cutoff of December 2023, and features a substantial 128k context length.

Key Capabilities

  • Multilingual Performance: Optimized for dialogue in English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai, with potential for fine-tuning in other languages.
  • Code Generation: Supports multilingual text and code output, demonstrating strong performance on benchmarks like HumanEval and MBPP++.
  • Instruction Following: Excels in assistant-like chat scenarios due to instruction tuning.
  • Tool Use: Shows significant improvements in tool-use benchmarks such as API-Bank and BFCL compared to its predecessor, Llama 3.
  • Reasoning and Math: Achieves high scores on reasoning benchmarks like ARC-C and mathematical tasks like GSM-8K and MATH.

Good for

  • Commercial and Research Use: Intended for a wide range of applications in both commercial products and academic research.
  • Multilingual Applications: Ideal for developing applications requiring robust performance across multiple languages.
  • Code-Related Tasks: Suitable for code generation and understanding.
  • Dialogue Systems: Optimized for creating assistant-like chat experiences.
  • Synthetic Data Generation: Can be used to generate synthetic data for improving other models.

Finetunes
4 models