whatmodel.sh / Qwen2.5 1.5B Instruct

Qwen2.5 1.5B Instruct

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

Qwen2.5 1.5B Instruct is an open-weights language model from Alibaba, released September 2024. Its published size is 1.54B parameters and a 33k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen2.5-1.5B-Instruct @989aa79: licence tag apache-2.0. 1,543,714,304 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-09: month the maker's Hugging Face repo was created (createdAt 2024-09-17). Ollama tags qwen2.5:1.5b-instruct-q4_K_M, qwen2.5:1.5b-instruct-q8_0 are the library's instruct/chat builds of this checkpoint (tag names on the pinned tags page); download sizes are estimated from the parameter count.

35.5±1.0#598 of 719 LLMsfrom 29 benchmark results across 12 sources

$ how it compares

The LLMs ranked either side of Qwen2.5 1.5B Instruct, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
690Llama 3 8B (base) Meta36±1local~6.6GB2024-04
691Mistral 7B v0.1 (base) Mistral AI36±0local~5.8GB2023-09
692Qwen2.5 1.5B Instruct Alibaba36±1local~2.4GB2024-09
693MPT-30B-Chat MosaicML35±2comparison only2023-06
694Gemma 7B (base) Google35±1local~7.1GB2024-02
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

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #263 of 300 * · 1.7% of its score · 1 test

Beats Magistral Small 1.0 (2506), Llama 4 Scout (109B-A17B), Phi-4 14B here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
BFCL v347.8% ✎▁▁▁▁▁▃▅█▅▄▁▁ beats 30/114medium · avg 52%▇ 1.7%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #356 of 458 · 9.5% of its score · 5 tests

Beats LFM2.5 8B-A1B, Qwen3-VL 2B Thinking, DeepSeek R1 Distill Qwen 7B here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
HumanEval30.5% ✎▁▁▁▁▁▃▅▅█▃▂▁ beats 37/190easy · avg 80%▇▇▇ 6.3%sharp · at its level · small run
CodeReviewQA: solution identification (hard), invariant accuracy37.8% ✎▃··▁▂▂▃▃█▃▄▂ beats 20/47medium · avg 54%▇ 1.2%sharp · noisy
CodeReviewQA: automated code refinement, exact match rate20.0% ✎▁▁▁▁▁▁▂▃▃█▅▄ beats 11/47hard · avg 33%▇ 1.1%typical
CodeReviewQA: change localisation (hard), invariant accuracy1.2% ✎█▃▁▁▂▂▂▃▅▅▅▇ beats 10/47medium · avg 36%▇ 0.8%sharp · too hard for it · noisy
CodeReviewQA: change type recognition, invariant accuracy78.2% ✎▁▁▁······▁▃█ beats 40/47easy · avg 67%▇ 0.0%blunt · at its level · noisy
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #375 of 393 · 6.7% of its score · 3 tests

Beats Phi-2 (2.7B) 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-controlled7.4% ✎▃▃·▁▁█▇▅·▁·▁ beats 1/22medium · avg 42%▇ 2.8%typical
IFEval prompt-level loose44.2% ✎▃▃▁▃▁▄▁▁█▃▃▁ beats 0/22easy · avg 80%▇ 2.3%at its level
IFEval (aggregation unspecified)47.4% ✎▁▁▃▁▂▄▄▆█▃▂▁ beats 7/141easy · avg 77%▇ 1.6%at its level · small run
math▁▂▃▅▆▅▄█▄▁▁▁ #341 of 398 · 23% of its score · 2 tests

Beats GPT-3.5 Turbo (0125), DBRX Instruct, Command R Plus (04-2024) here, though they rank higher overall. Trails Qwen3 0.6B, Gemma 7B (base) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K +166.2% ✎▂▃··▃▂▅█▆▃▂▃ beats 3/22very easy · avg 88%▇▇▇▇▇▇ 16%sharp · at its level · several runs
MATH (subset unspecified) +140.6% ✎▃▃█▂▂▆·····▁ beats 9/22medium · avg 62%▇▇▇ 7.4%at its level · small run · several runs
other tests15 not in a task · 59% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard +145.8% ✎▁▄·▃▁·▅█·▄▄▁ beats 4/22easy · avg 75%▇▇▇▇▇▇▇ 16%sharp · at its level · several runs
MMLU59.7% ✎▁▂·▁·▃▁█·▃▃▅ beats 4/22easy · avg 77%▇▇▇▇▇ 11%at its level
CPTUB average (SpeakLeash, 0-5 judge scale)35.2% ✎▁▃▂▂▃█▃▆▅▄▂▃ beats 4/95easy · avg 66%▇▇ 6.2%at its level
Polish EQ-Bench v2 (SpeakLeash)27.8% ✎▁··▁▁▃▂▆▅▆█▆ beats 6/70medium · avg 64%▇▇ 4.4%at its level
Open Persian LLM Leaderboard v2 average (21 tasks)36.8% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 9/81medium · avg 58%▇ 3.5%blunt · at its level
BRIDGE clinical text, zero-shot average (87 tasks)22.2% ✎▁▁▁▁▁▃▂▃▄▆█▃ beats 15/97hard · avg 32%▇ 3.1%blunt
HellaSwag60.9% ✎▁▁▁▁▂▂▃▇█▆▅▃ beats 22/85easy · avg 86%▇ 2.6%at its level · small run
TruthfulQA MC246.5% ✎▃▃▂·▃▃·▅█▂▅▂ beats 3/21medium · avg 63%▇ 1.9%blunt · at its level
7 more tests
testresultvs the others in its rundifficultycountswhy
AraGen-v3 3C3H score (Inception, Claude Sonnet 4 judge)13.2% ✎▁▁·▁▂▃▄█▄▅▄▄ beats 9/66hard · avg 31%▇ 1.7%typical
Open PT-LLM Leaderboard average (9 Portuguese tasks)65.6% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 117/276medium · avg 73%▇ 1.6%at its level · noisy
PopQA15.5% ✎▁▁▁▃▃▃▁█▁▃·▄ beats 3/21hard · avg 32%▇ 1.5%blunt
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy76.0% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 52/146easy · avg 83%▇ 1.3%at its level · noisy
PIQA73.2% ✎▁▁▁▃▃▄▇█▄▃▁▁ beats 26/83easy · avg 86%▇ 1.2%blunt · at its level · small run
EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy64.6% ✎▃·▁▁▂▁▂▅█▇▄▁ beats 40/100medium · avg 68%▇ 1.1%noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)84.2% ✎▁▁▃▃▄▄█▆▃▂▂▁ beats 12/140very easy · avg 94%▇ 0.9%blunt · 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 1.5B Instruct · needs ~1.1 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    llama cli -hf Qwen/Qwen2.5-1.5B-Instruct-GGUF:q4_k_m
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf Qwen/Qwen2.5-1.5B-Instruct-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 1.5B Instruct · needs ~1.1 GB · from huggingface.co/Qwen (official)

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct-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 1.5B Instruct · needs ~1.42 GB · from ollama.com
    chat in the terminal
    ollama run qwen2.5:1.5b-instruct-q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Qwen2.5 1.5B Instruct · needs ~0.9 GB, Mac only · from huggingface.co/lmstudio-community
    chat in the terminal
    mlx_lm.chat --model lmstudio-community/Qwen2.5-1.5B-Instruct-MLX-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model lmstudio-community/Qwen2.5-1.5B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · 2 OLMo 2 Furious (Team OLMo,… 41%, SmolLM2 paper (Hugging Face)… 29%, SpeakLeash CPTUB (Complex Pol… 6% and 9 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
2 OLMo 2 Furious (Team OLMo, Ai2) — authors' own evaluations of OLMo 2 Instruct and peer instruct models8▇▇▇▇▇▇▇▇▇▇▇▇ 41%CC BY 4.0
SmolLM2 paper (Hugging Face) — authors' own instruct-model evaluations7▇▇▇▇▇▇▇▇▇ 29%CC BY 4.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇▇ 6%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇ 4%Apache-2.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇ 4%Apache-2.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)4▇ 3%CC-BY-4.0
BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab)1▇ 3%MIT
EuroEval leaderboards (results collection), English generative leaderboard2▇ 2%MIT
Berkeley Function Calling Leaderboard result archive1▇ 2%Apache-2.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)1▇ 2%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 2%Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 1%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 35.5

$ 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