whatmodel.sh / Tulu 2 DPO 7B

Tulu 2 DPO 7B

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

Tulu 2 DPO 7B is an open-weights language model from Allen Institute for AI, released November 2023. Its published size is 7B parameters. Licence: AI2 ImpACT Low-risk; Llama 2 base terms.

Catalogue identity notes

Exact full-precision model evaluated in the Tulu 2 paper (arXiv:2311.10702v2). QLoRA ablations are separate and not imported.

Model information sources

Release date ↗ · Weights licence ↗

35.1±1.4#600 of 719 LLMsfrom 8 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of Tulu 2 DPO 7B, out of 828 with a score, then more from Allen Institute for AI.

#modelscorewhereRAMreleased
693MPT-30B-Chat MosaicML35±2comparison only2023-06
694Gemma 7B (base) Google35±1local~7.1GB2024-02
695Tulu 2 DPO 7B Allen Institute for AI35±1local~5.8GB2023-11
696Eurus-2-7B-SFT PRIME-RL35±3local~6.4GB2024-12
697Falcon 40B (base) TII35±1local~27GB2023-05
more from Allen Institute for AI
493OLMo 2 13B Instruct (1124) Allen Institute for AI46±1local~9.6GB2024-12
516Tulu 2 70B Allen Institute for AI46±2local~28GB2023-11
557Code Tulu 2 34B Allen Institute for AI44±2local~72GB2023-11

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▃▃▆██▇▆▄▃▁ #441 of 458 * · 15% of its score · 1 test

Beats Molmo 7B-D (0924), DeepSeek R1 Distill Qwen 1.5B, Tulu 2 7B here, though they rank higher overall. Trails Falcon3 1B Instruct, Granite 3.0 1B-A400M Instruct, Gemma 1.1 2B IT here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Tulu 2: CodexEval pass@10 lab40.0% ✎▃▃▃·▆▃▃█▃▂▂▅ beats 5/25very easy · avg 86%▇▇▇▇▇▇ 15%sharp · at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #362 of 393 * · 38% of its score · 2 tests

Beats Gemma 4 E2B, Code Llama 70B Instruct, Phi-3-mini-128k-instruct (Apr 2024) 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-controlled9.2% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 20/122medium · avg 39%▇▇▇▇▇▇▇▇▇▇▇▇ 30%typical
AlpacaEval 1.0 win rate (vs text-davinci-003, GPT-4 judge)84.2% ✎▂▂·▆·▇▃▆█▆▄▂ beats 22/45very easy · avg 96%▇▇▇ 7.9%sharp · noisy
math▁▂▃▅▆▅▄█▄▁▁▁ #347 of 398 * · 8.3% of its score · 1 test

Beats Command R Plus (04-2024), Phi-3-mini-4k-instruct (June 2024 update), Yi 34B Chat here, though they rank higher overall. Trails Qwen3 0.6B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K lab34.5% ✎▆▂▅█▅▅▃▆···▂ beats 11/25easy · avg 73%▇▇▇ 8.3%sharp · at its level
other tests4 not in a task · 38% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard lab45.5% ✎▁·····▂█▂▅·▁ beats 6/25easy · avg 77%▇▇▇▇▇▇▇▇▇ 23%sharp · at its level
MMLU lab50.7% ✎▁▁▃▃▄█▄▁█··▁ beats 10/25easy · avg 74%▇▇▇ 8.7%at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)63.9% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 106/276medium · avg 73%▇ 3.3%at its level · noisy
Tulu 2: TyDiQA gold passage, 1-shot F1 lab44.5% ✎▂··▁▃▂▃▂█▅▁▂ beats 9/25medium · avg 61%▇ 3.1%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 Tulu 2 DPO 7B · needs ~4.1 GB · from huggingface.co/TheBloke
    chat in the terminal
    llama cli -hf TheBloke/tulu-2-dpo-7B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf TheBloke/tulu-2-dpo-7B-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 Tulu 2 DPO 7B · 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/tulu-2-dpo-7B-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 Tulu 2 DPO 7B · needs ~4.1 GB · from huggingface.co/TheBloke
    chat in the terminal
    ollama run hf.co/TheBloke/tulu-2-dpo-7B-GGUF:Q4_K_M

Other ways to run it

  1. Run it as published · needs ~13.48 GB · from huggingface.co/allenai
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='allenai/tulu-2-dpo-7b', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Weights and documentation: huggingface.co/allenai ↗

$ how the score was worked out · Tulu 2 — base and instruction… 58%, AlpacaEval 1.0 and 2.0 leader… 38%, Open Portuguese LLM Leaderboa… 3%

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
Tulu 2 — base and instruction-tuned evaluations · lab-reported5▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 58%CC BY 4.0
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)2▇▇▇▇▇▇▇▇▇▇▇ 38%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 3%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 35.1

$ 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