whatmodel.sh / Qwen2.5-Omni 7B

Qwen2.5-Omni 7B

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

Qwen2.5-Omni 7B is an open-weights language model from Alibaba, released March 2025. Its published size is 10.7B parameters. Licence: apache-2.0.

Catalogue identity notes

Comparison-only. Qwen's omni-modal model, repo Qwen/Qwen2.5-Omni-7B @ae9e169 (licence apache-2.0 via license_name/link, created 2025-03-22); AHELM identifier 'qwen2.5-omni-7b'. No verified run command recorded.

Model information sources

· Parameter count ↗

10,732,225,440 parameters in the repo's safetensors metadata (AHELM Table 3: 10.7B, incl. audio/vision encoders and speech decoder).

57.1±2.4#235 of 719 LLMsfrom 10 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of Qwen2.5-Omni 7B, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
279Gemini 2.5 Flash-Lite Google57±2api $0.17/M2025-07
280Qwen2.5-Max (2025-01-25) Alibaba57±1comparison only2025-01
281Qwen2.5-Omni 7B Alibaba57±2local~6.4GB2025-03
282GPT-4o (2024-05-13) OpenAI57±0api $7.50/M2024-05
283QED-Nano LM Provers57±2local~4.1GB2026-02
more from Alibaba
18Qwen3.8 Max (2.4T-A95B) Alibaba78±1api $3.00/M +local~380GB2026-08
34Qwen3.8 Max (0902) Alibaba76±1api $3.00/M2026-09
37Qwen3.8-Flash-Next Alibaba75±1local~59GB2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests10 not in a task · 98% of its score
testresultvs the others in its rundifficultycountswhy
AHELM Speech Robust Bench (LibriSpeech-clean with noise) transcription WER10.3% ✎▂·█··▆▄·▃▂·▄ beats 10/13avg 33%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 47%typical
AHELM CoRe-Bench conversational audio reasoning (pseudo-exact match)56.0% ✎▂·▄··▂▃·▄▂█▄ beats 4/13medium · avg 53%▇▇▇▇▇▇ 15%at its level
AHELM VoxCeleb2 same-speaker detection (exact match)58.1% ✎▂·▂·▂█▂▄▄··▄ beats 10/13very hard · avg 54%▇▇▇▇ 10%sharp · noisy
AHELM LibriSpeech transcription by audio LLMs (WER)10.3% ✎▃···▃·█▃▅▃▅█ beats 4/13avg 11%▇▇▇▇ 10%blunt
AHELM MUStARD audio sarcasm detection (exact match)58.8% ✎▂·······▆█▆▆ beats 7/13very hard · avg 57%▇▇▇ 7.5%blunt
AHELM AIR-Bench Foundation music subsets (exact match)74.3% ✎▃█▃▅▃·▃·▃▅▃▃ beats 13/13medium · avg 57%▇▇ 4.0%blunt · at its level
AHELM MELD audio emotion classification (pseudo-exact match)49.1% ✎▃··▃▃▃·▃▅▅█▅ beats 11/13medium · avg 40%▇ 2.5%blunt · at its level
AHELM FLEURS 7-language transcription by audio LLMs (WER)193.2% ✎▄▃···▂▆▂█▂·▂ beats 2/13avg 80%▇ 1.5%blunt · noisy
2 more tests
testresultvs the others in its rundifficultycountswhy
AHELM Multilingual LibriSpeech 7-language transcription by audio LLMs (WER)41.6% ✎▂▃▂··▂▂·▂▂█▆ beats 1/13avg 13%▇ 0.2%blunt · noisy
AHELM VocalSound human-sound recognition (prefix exact match)90.4% ✎██▄▄·█·█·█▄▄ beats 13/13easy · avg 67%▇ 0.1%blunt · 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 Qwen2.5-Omni 7B · needs ~4.7 GB · from huggingface.co/ggml-org
    chat in the terminal
    llama cli -hf ggml-org/Qwen2.5-Omni-7B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf ggml-org/Qwen2.5-Omni-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 Qwen2.5-Omni 7B · needs ~4.7 GB · from huggingface.co/ggml-org

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/ggml-org/Qwen2.5-Omni-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 Qwen2.5-Omni 7B · needs ~4.7 GB · from huggingface.co/ggml-org
    chat in the terminal
    ollama run hf.co/ggml-org/Qwen2.5-Omni-7B-GGUF:Q4_K_M

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · AHELM: A Holistic Evaluation… 98%

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
AHELM: A Holistic Evaluation of Audio-Language Models (Stanford CRFM, arXiv:2508.21376v2)10▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 98%CC BY 4.0
prior · pull toward the middle▇ 2%
weighted average= 57.1

$ 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