TīrthaModels

Models / Meta

Llama-3.1-8B-Instruct

Gets 80 of every 100 tool calls right on the public test. Best at calling a single tool, weaker at choosing the right tool from several.

Open weights · Llama 3.1 Community Licenseby MetaReleased Jul 18, 2024Built with LlamaTested on Tīrtha Bench

Llama-3.1-8B-Instruct is made by Meta. It takes text in and answers in text, with a context window of 131,072 tokens.

Start using Llama-3.1-8B-Instruct in 3 steps

Read the docs
  1. Get an API keyFree to create. It works for every model in the catalogue.Get an API key
  2. Add creditsPay as you go. Credits work on every model.Add credits
  3. Point your client at TīrthaAny OpenAI-compatible client works. Use your key and this base URL, and name the model.
    base_url: "https://api.tirtha.ai/v1" api_key: "YOUR_TIRTHA_KEY" model: "catalog/llama-3.1-8b-instruct"

At a glance

MakerMeta
Acceptstext
Context131,072 tokens
Max output16,384 tokens
LicenceLlama 3.1 Community License
ReleasedJul 18, 2024

Tested on Tīrtha Bench

Tool calling (public BFCL v4 test)

Overall: 80.5%*. Picks the right tool and fills in the inputs correctly on 80 of every 100 tasks.

One tool 84% · Picking the right tool 76% · Several calls at once 78%

A task counts as right only if the model calls the right tool with the right inputs. 800 tasks. Run on Oct 29, 2026. *Measured by Spanda Works on third-party hosts, not through Tīrtha. Results through Tīrtha may differ.

Handled several calls at once in our tests.

Measured as llama-3.1-8b-instruct-turbo.

See Llama-3.1-8B-Instruct beside every other model on Tīrtha Bench →

What it supports

From the maker

Facts here are read from the maker's published card; claims and benchmarks on that card are the maker's own, not measured by Tīrtha.

Licence

Weights: Llama 3.1 Community License. Llama 3.1 Community License. Llama 3.1 Acceptable Use Policy.

Built with Llama

Llama 3.1 is licensed under the Llama 3.1 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.