whatmodel.sh / Vicuna 7B v1.5

Vicuna 7B v1.5

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

Vicuna 7B v1.5 is an open-weights language model from LMSYS, released August 2023. Its published size is 7B parameters and a 4k-token context window. Licence: llama2.

Catalogue identity notes

Arena entry 'vicuna-7b' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'vicuna-7b' as 'Vicuna-7B' (LMSYS, Llama 2 Community), linking https://huggingface.co/lmsys/vicuna-7b-v1.5. Maker repo lmsys/vicuna-7b-v1.5 @3321f76: licence tag llama2. Parameters: nominal size from the card/model name (fine-tuned Llama 2 7B). Released 2023-08: FastChat README (pinned 2c68a13): '[2023/08] We released Vicuna v1.5 based on Llama 2'. The Arena key 'vicuna-7b' is unversioned; the dataset's licence field ('Llama 2 Community') and LMArena's own leaderboard table (link lmsys/vicuna-7b-v1.5) identify the Llama 2-based v1.5 checkpoint.

32.8±1.2#619 of 719 LLMsfrom 10 benchmark results across 7 sources

$ how it compares

The LLMs ranked either side of Vicuna 7B v1.5, out of 828 with a score, then more from LMSYS.

#modelscorewhereRAMreleased
714GPT-3 davinci (API base model) OpenAI33±1comparison only2020-06
715StarCoderBase 15.5B BigCode33±1local~12GB2023-05
716Vicuna 7B v1.5 LMSYS33±1local~5.8GB2023-08
717Guanaco 33B University of Washington33±2local~70GB2023-05
718Stable Code 3B Stability AI33±1local~3.3GB2024-01
more from LMSYS
635Vicuna 33B v1.3 LMSYS39±1local~23GB2023-06
679Vicuna 13B v1.5 LMSYS36±2local~9.8GB2023-08
804FastChat-T5 3B LMSYS17±2local~8.5GB2023-04

see all LLMs · find the ones that fit your machine

$ what it's good at

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #371 of 393 · 52% of its score · 4 tests

Beats Phi-2 (2.7B), Qwen2.5 1.5B Instruct, Nous-Hermes-13b here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct, Granite 4.0 350M here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AlpacaEval 2 length-controlled7.6% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 16/122medium · avg 39%▇▇▇▇▇▇▇ 17%typical
Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024)1058.0 ✎▂▁▂▁▃▇█▆▇▄▂▂ beats 29/95avg 1201▇▇▇▇▇ 12%blunt
LMArena (text)1115.2 ✎▁▁▁▃▃▃▃▆▆▇█▃ beats 20/379avg 1268▇▇▇▇▇ 12%blunt
MT-Bench (GPT-4 single-answer grading, LMSYS runs) lab6.1700 MT-Bench score (1-10) ✎▆▂▂▆▆▄██▂▄▄▆ beats 13/28avg 8.819 MT-Bench score (1-10)▇▇▇▇ 10%sharp · noisy
other tests6 not in a task · 48% of its score
testresultvs the others in its rundifficultycountswhy
LiveBench 2024-07-2612.6% ✎▁▁▁▃▂▃▂█▆▃▅▃ beats 7/61medium · avg 41%▇▇▇▇▇▇▇▇▇▇ 25%typical
Natural Questions (closed book)20.6% ✎▁·▁·▁▂▁▂·▅▃█ beats 16/28medium · avg 58%▇▇▇▇▇▇▇ 17%typical
TriviaQA56.4% ✎▁··▁█▃▃▁▄▅▄█ beats 21/28easy · avg 79%▇ 2.1%at its level · noisy
Open PT-LLM Leaderboard average (9 Portuguese tasks)57.0% ✎▃▁▁▁▁▁▁▁▃█▅▁ beats 82/276medium · avg 73%▇ 2.1%at its level · noisy
HaluEval QA hallucination detection (accuracy)38.1% ✎█▁▁·▁▁·▁▁▁▁▁ beats 6/27medium · avg 70%▇ 1.1%sharp · noisy
TruthfulQA MC250.4% ✎▅▃█▆▁▃▅···▁▁ beats 23/31medium · avg 64%▇ 1.0%blunt · at its level

$ 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 Vicuna 7B v1.5 · needs ~4.1 GB · from huggingface.co/TheBloke
    chat in the terminal
    llama cli -hf TheBloke/vicuna-7B-v1.5-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf TheBloke/vicuna-7B-v1.5-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 Vicuna 7B v1.5 · needs ~4.1 GB · from huggingface.co/TheBloke

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/TheBloke/vicuna-7B-v1.5-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Vicuna 7B v1.5 · needs ~4.7 GB · from ollama.com
    chat in the terminal
    ollama run vicuna:7b-v1.5-q4_K_M

Official weights and documentation: huggingface.co/lmsys ↗

$ how the score was worked out · LiveBench model judgments (HF… 25%, Hallucinations Leaderboard re… 21%, AlpacaEval 1.0 and 2.0 leader… 17% and 4 more

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
LiveBench model judgments (HF livebench/model_judgment, 2024-10-22 revision) with the LiveBench question datasets1▇▇▇▇▇▇▇ 25%Apache-2.0 (stated in the LiveBench datasheet)
Hallucinations Leaderboard results (HF dataset hallucinations-leaderboard/results)4▇▇▇▇▇▇ 21%Apache-2.0
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)1▇▇▇▇▇ 17%Apache-2.0
Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3)1▇▇▇▇ 12%CC-BY-4.0
LMArena leaderboard dataset1▇▇▇▇ 12%CC BY 4.0
LMSYS Chatbot Arena leaderboard Space: MT-Bench column of the published leaderboard table · lab-reported1▇▇▇ 10%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 2%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 32.8

$ 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