whatmodel.sh / Llama-3-70B-Instruct-Gradient-1048k

Llama-3-70B-Instruct-Gradient-1048k

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Llama-3-70B-Instruct-Gradient-1048k is an open-weights language model from Gradient AI, released May 2024. Its published size is 70.55B parameters. Licence: llama3.

Catalogue identity notes

Maker repo gradientai/Llama-3-70B-Instruct-Gradient-1048k @51bcb55: licence tag llama3. 70,553,706,496 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-05: month the maker's Hugging Face repo was created (createdAt 2024-05-03).

49.5±5.7#459 of 949 LLMsfrom 2 benchmark results across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of Llama-3-70B-Instruct-Gradient-1048k, out of 1132 with a score, then more from Gradient AI.

#modelscorewhereRAMreleased
590Nova Pro Amazon50±1api $1.40/M2024-12
591WizardCoder 33B V1.1 WizardLM50±1local~22GB2024-01
592Llama-3-70B-Instruct-Gradient-1048k Gradient AI50±6local~29GB2024-05
593Mistral Small Instruct 22B (2409) Mistral AI49±2local~15GB2024-09
594Neural Chat 7B v3.3 Intel49±6local~6.1GB2023-12
more from Gradient AI
674Llama-3-8B-Instruct-Gradient-1048k Gradient AI47±5local~6.2GB2024-04

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests2 not in a task · 86% of its score
testresultvs the others in its rundifficultycountswhy
RULER 64k80.9% ✎▁▁·▁·▂▃▄█▂·▁ beats 8/22easy · avg 83%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 62%typical
RULER 128k72.1% ✎▄··▂▄▂▆██▄▄▂ beats 12/21easy · avg 68%▇▇▇▇▇▇▇▇▇▇ 24%at its level · noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Llama-3-70B-Instruct-Gradient-1048k · needs ~42.5 GB · from huggingface.co/featherless-ai-quants
    chat in the terminal
    llama cli -hf featherless-ai-quants/gradientai-Llama-3-70B-Instruct-Gradient-1048k-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf featherless-ai-quants/gradientai-Llama-3-70B-Instruct-Gradient-1048k-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Llama-3-70B-Instruct-Gradient-1048k · needs ~42.5 GB · from huggingface.co/featherless-ai-quants

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/featherless-ai-quants/gradientai-Llama-3-70B-Instruct-Gradient-1048k-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Llama-3-70B-Instruct-Gradient-1048k · needs ~39.7 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Llama-3-70B-Instruct-Gradient-1048k-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Llama-3-70B-Instruct-Gradient-1048k-4bit

Other ways to run it

  1. Run it as published · needs ~141.11 GB · from huggingface.co/gradientai (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='gradientai/Llama-3-70B-Instruct-Gradient-1048k', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/gradientai ↗

$ how the score was worked out · NVIDIA RULER repository READM… 86%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
NVIDIA RULER repository README results table2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 86%Apache-2.0
prior · pull toward the middle▇▇▇▇ 14%
weighted average= 49.5

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json