whatmodel.sh / Starling-LM 7B alpha

Starling-LM 7B alpha

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

Starling-LM 7B alpha is an open-weights language model from UC Berkeley (NEST), released November 2023. Its published size is 7.24B parameters and a 8k-token context window. Licence: apache-2.0.

Catalogue identity notes

Arena entry 'starling-lm-7b-alpha' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'starling-lm-7b-alpha' as 'Starling-LM-7B-alpha' (UC Berkeley, CC-BY-NC-4.0), linking https://huggingface.co/berkeley-nest/Starling-LM-7B-alpha. Maker repo berkeley-nest/Starling-LM-7B-alpha @1dddf3b: licence tag apache-2.0. 7,241,748,480 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2023-11: month the maker's Hugging Face repo was created (createdAt 2023-11-25). Card: 'Apache-2.0 license under the condition that the model is not used to compete with OpenAI'; also described as a non-commercial research preview (the Arena licence field says CC-BY-NC-4.0).

38.9±1.1#545 of 719 LLMsfrom 11 benchmark results across 8 sources

$ how it compares

The LLMs ranked either side of Starling-LM 7B alpha, out of 828 with a score.

#modelscorewhereRAMreleased
635Vicuna 33B v1.3 LMSYS39±1local~23GB2023-06
636DeepSeek-Coder 6.7B Base DeepSeek39±1local~5.8GB2023-10
637Starling-LM 7B alpha UC Berkeley (NEST)39±1local~6.1GB2023-11
638Granite 8B Code Base (4K) IBM39±1local~6.6GB2024-04
639StarCoder2 15B BigCode39±1local~11GB2024-02

see all LLMs · find the ones that fit your machine

$ what it's good at

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #337 of 393 · 60% of its score · 4 tests

Beats OpenThinker3 7B, Gemma 4 E2B, Code Llama 70B Instruct here, though they rank higher overall. Trails Granite 4.0 H 350M, LFM2 350M, Gemma 3 1B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AlpacaEval 2 length-controlled14.7% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 37/122medium · avg 39%▇▇▇▇▇▇▇▇▇▇ 25%typical
Arena-Hard (v0.1, GPT-4-1106 judge)12.8% ✎▂▁▁▅█▅▄▃▃▅▃▁ beats 7/60medium · avg 52%▇▇▇▇▇▇ 14%sharp
Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024)1099.0 ✎▂▁▂▁▃▇█▆▇▄▂▂ beats 44/95avg 1201▇▇▇▇ 11%blunt
LMArena (text)1167.4 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 48/379avg 1268▇▇▇▇ 11%blunt
other tests7 not in a task · 39% of its score
testresultvs the others in its rundifficultycountswhy
Natural Questions (closed book)28.2% ✎▁·▁·▁▂▁▂·▅▂█ beats 25/28medium · avg 58%▇▇▇▇▇▇▇ 19%at its level
CPTUB average (SpeakLeash, 0-5 judge scale)52.6% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 18/95easy · avg 66%▇▇▇ 8.0%at its level
Polish EQ-Bench v2 (SpeakLeash)46.3% ✎▁··▁▁▂▂▆▅▆█▆ beats 19/70medium · avg 64%▇▇ 6.0%at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)69.6% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 162/276medium · avg 73%▇ 2.2%at its level · noisy
TriviaQA66.0% ✎▁··▁█▃▃▁▄▅▄█ beats 26/28easy · avg 79%▇ 1.9%at its level · noisy
HaluEval QA hallucination detection (accuracy)59.7% ✎█▁▁·▁▁·▁▁▃▁▁ beats 24/27medium · avg 70%▇ 1.6%sharp · noisy
TruthfulQA MC247.3% ✎▅▃█▆▃▂▅···▁▁ beats 22/31medium · avg 64%▇ 0.9%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 Starling-LM 7B alpha · needs ~4.4 GB · from huggingface.co/MaziyarPanahi
    chat in the terminal
    llama cli -hf MaziyarPanahi/Starling-LM-7B-alpha-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf MaziyarPanahi/Starling-LM-7B-alpha-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 Starling-LM 7B alpha · needs ~4.4 GB · from huggingface.co/MaziyarPanahi

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/MaziyarPanahi/Starling-LM-7B-alpha-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 Starling-LM 7B alpha · needs ~4.84 GB · from ollama.com
    chat in the terminal
    ollama run starling-lm:7b-alpha-q4_K_M

Official weights and documentation: huggingface.co/berkeley-nest ↗

$ how the score was worked out · AlpacaEval 1.0 and 2.0 leader… 25%, Hallucinations Leaderboard re… 23%, Arena-Hard-Auto v0.1 leaderbo… 14% and 5 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
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)1▇▇▇▇▇▇▇▇ 25%Apache-2.0
Hallucinations Leaderboard results (HF dataset hallucinations-leaderboard/results)4▇▇▇▇▇▇▇ 23%Apache-2.0
Arena-Hard-Auto v0.1 leaderboard (repository README, pre-v2)1▇▇▇▇ 14%Apache-2.0
Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3)1▇▇▇ 11%CC-BY-4.0
LMArena leaderboard dataset1▇▇▇ 11%CC BY 4.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇▇ 8%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇▇ 6%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 2%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 38.9

$ 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