whatmodel.sh / Qwen1.5 72B Chat

Qwen1.5 72B Chat

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

Qwen1.5 72B Chat is an open-weights language model from Alibaba, released February 2024. Its published size is 72.29B parameters and a 32k-token context window. Licence: tongyi-qianwen.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2024-02-04). Chat checkpoint Qwen1.5-72B-Chat. 72,287,920,128 parameters. ollama qwen:72b-chat = 72b-chat-v1.5-q4_0 (same digest).

45.4±0.5#437 of 719 LLMsfrom 19 benchmark results across 15 sources

$ how it compares

The LLMs ranked either side of Qwen1.5 72B Chat, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
520InternLM2.5 20B Chat Shanghai AI Laboratory (InternLM)46±2local~14GB2024-08
521StarCoder2-15B-Instruct-v0.1 BigCode46±1local~11GB2024-04
522Qwen1.5 72B Chat Alibaba45±1local~31GB2024-02
523speechless-starcoder2-15b uukuguy45±1local~12GB2024-03
524Ministral 3 3B Mistral45±1api $0.10/M +local~3.5GB2025-12
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

coding▁▁▁▃▆██▇▆▄▃▁ #323 of 458 · 47% of its score · 3 tests

Beats GLM-4.7-Flash, Qwen3-VL 8B Instruct, Qwen3-VL 4B Instruct here, though they rank higher overall. Trails Phi-3-mini-128k-instruct (Apr 2024), DeepSeek LLM 67B Chat, WizardLM 70B V1.0 here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval +168.3% ✎▁·▁▁▃▄▆▅█▃▁▁ beats 54/99easy · avg 79%▇▇▇▇▇▇▇▇▇▇▇▇ 30%sharp · at its level · several runs
HumanEval+59.1% ✎▁·▁▁▁▃▄▄▄█▃▁ beats 50/99easy · avg 74%▇▇▇▇▇▇ 15%sharp · at its level
MBPP+61.6% ✎▁···▁▁▂▅▇█▆▂ beats 40/75easy · avg 68%▇ 2.7%at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #287 of 393 · 20% of its score · 5 tests

Beats Ministral 3 14B Reasoning 2512, Gemma 4 E4B, OpenThinker3 7B here, though they rank higher overall. Trails Flan-T5 XXL, LFM2 700M, EXAONE 3.5 2.4B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AlpacaEval 2 length-controlled36.6% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 92/122medium · avg 39%▇▇▇ 8.0%at its level
Arena-Hard (v0.1, GPT-4-1106 judge)36.1% ✎▂▁▁▅█▅▄▃▃▅▃▁ beats 30/60medium · avg 52%▇▇ 5.6%sharp · at its level
Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024)1148.0 ✎▂▁▂▁▃▇█▆▇▄▂▂ beats 65/95avg 1201▇ 2.3%blunt
LMArena (text)1233.5 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 85/379avg 1268▇ 2.2%blunt
Arabic IFEval prompt-level (Inception)14.1% ✎▁·▁▁▂▁▄▄▆▄█▄ beats 8/38medium · avg 49%▇ 1.8%sharp · at its level
other tests10 not in a task · 33% of its score
testresultvs the others in its rundifficultycountswhy
LiveBench 2024-07-2629.3% ✎▁▁▁▂▂▃▂█▆▃▅▃ beats 25/61medium · avg 41%▇▇▇ 8.7%at its level
EvoEval (overall)42.0% ✎▂▂▃▄▄█▁▄▅▃▁▃ beats 28/42medium · avg 55%▇▇▇ 7.1%at its level
MMLU-Pro47.1% ✎▃▂▅█▃·█▃▅··▂ beats 11/23medium · avg 57%▇▇ 3.8%at its level
BigCodeBench (Complete) lab39.8% ✎▂▁▁▄▂▅▇█▆▂▂▂ beats 27/58medium · avg 50%▇ 2.7%at its level
BigCodeBench (Instruct) lab32.6% ✎▃▂·▆▆██▆▅▃▃▃ beats 11/33medium · avg 41%▇ 2.7%at its level
WildBench v2 WB-Score (task macro)39.9277 WB-Score ✎▁·▁▁·▃█▄▇▆▅▆ beats 30/52avg 45.199 WB-Score▇ 2.6%sharp
GPQA Diamond28.8% ✎▁▁▁▃▂▄▆▄▅▇█▅ beats 15/208hard · avg 40%▇ 1.7%sharp
CPTUB average (SpeakLeash, 0-5 judge scale)63.2% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 42/95easy · avg 66%▇ 1.6%at its level
2 more tests
testresultvs the others in its rundifficultycountswhy
Polish EQ-Bench v2 (SpeakLeash)68.0% ✎▁··▁▁▂▂▆▅▆█▆ beats 47/70medium · avg 64%▇ 1.3%at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)74.3% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 217/276medium · avg 73%▇ 0.5%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 Qwen1.5 72B Chat · needs ~44.1 GB · from huggingface.co/mradermacher
    chat in the terminal
    llama cli -hf mradermacher/Qwen1.5-72B-Chat-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf mradermacher/Qwen1.5-72B-Chat-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 Qwen1.5 72B Chat · needs ~44.1 GB · from huggingface.co/mradermacher

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-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 Qwen1.5 72B Chat · needs ~43.87 GB · from ollama.com
    chat in the terminal
    ollama run qwen:72b-chat

MLX (Mac)

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Qwen1.5 72B Chat · needs ~42.5 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Qwen1.5-72B-Chat-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Qwen1.5-72B-Chat-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · EvalPlus / EvalPerf published… 37%, Top Leaderboard Ranking = Top… 18%, LiveBench model judgments (HF… 9% and 12 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
EvalPlus / EvalPerf published evaluation runs3▇▇▇▇▇▇▇▇▇▇▇ 37%Apache-2.0
Top Leaderboard Ranking = Top Coding Proficiency, Always? EvoEval: Evolving Coding Benchmarks via LLM — authors' own runs2▇▇▇▇▇ 18%CC BY 4.0
LiveBench model judgments (HF livebench/model_judgment, 2024-10-22 revision) with the LiveBench question datasets1▇▇▇ 9%Apache-2.0 (stated in the LiveBench datasheet)
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)1▇▇ 8%Apache-2.0
Arena-Hard-Auto v0.1 leaderboard (repository README, pre-v2)1▇▇ 6%Apache-2.0
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v4 · lab-reported2▇▇ 5%CC BY 4.0
MMLU-Pro author-released evaluation runs1▇ 4%Apache-2.0
WildBench v2 leaderboard data (Ai2)1▇ 3%Apache-2.0
Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3)1▇ 2%CC-BY-4.0
LMArena leaderboard dataset1▇ 2%CC BY 4.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)1▇ 2%Apache-2.0
Epoch AI — internal evaluations1▇ 2%CC BY 4.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇ 2%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇ 1%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 1%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 45.4

$ 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