whatmodel.sh / Qwen2.5 Coder 14B Instruct

Qwen2.5 Coder 14B Instruct

+ contribute

Qwen2.5 Coder 14B Instruct is an open-weights language model from Alibaba, released November 2024. Its published size is 14.77B parameters and a 131k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen2.5-Coder-14B-Instruct @aedcc2d: licence tag apache-2.0. 14,770,033,664 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-11: month the maker's Hugging Face repo was created (createdAt 2024-11-06). Ollama tags qwen2.5-coder:14b-instruct-q4_K_M, qwen2.5-coder:14b-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.

55.1±1.2#264 of 719 LLMsfrom 15 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of Qwen2.5 Coder 14B Instruct, out of 828 with a score, then more from Alibaba.

#	model	score	where	RAM	released
316	Gemini 1.5 Pro Experimental (0801) Google	55±1	comparison only		2024-08
317	Qwen3 4B Instruct 2507 Alibaba	55±1	local	~3.9GB	2025-08
318	Qwen2.5 Coder 14B Instruct Alibaba	55±1	local	~10GB	2024-11
319	Qwen3-VL 8B Instruct Alibaba	55±1	api $0.31/M +local	~6.8GB	2025-10
320	Spark-X2.5 1.7B iFLYTEK (SparkLLM)	55±1	local	~2.7GB	2026-08
	more from Alibaba
18	Qwen3.8 Max (2.4T-A95B) Alibaba	78±1	api $3.00/M +local	~380GB	2026-08
34	Qwen3.8 Max (0902) Alibaba	76±1	api $3.00/M		2026-09
37	Qwen3.8-Flash-Next Alibaba	75±1	local	~59GB	2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #211 of 458 · 83% of its score · 12 tests

Beats Qwen3.5 4B, Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B) here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, DeepSeek-Coder-V2 Instruct (0724) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Aider code editing (133 Exercism Python exercises, two attempts)	69.2% ✎	▁·▁▂▄▁▆▄█▆▂▂ beats 46/62	medium · avg 54%	▇▇▇▇▇▇▇▇▇ 23%	at its level
HumanEval lab	89.6% ✎	▄·▄···█··█·▄ beats 5/6	very easy · avg 86%	▇▇▇▇▇ 12%	sharp · maxed out
HumanEval+ lab	87.2% ✎	▄·▄····▄▄·▄█ beats 6/6	easy · avg 82%	▇▇▇▇ 11%	sharp
MBPP lab	86.2% ✎	█···███·██·█ beats 5/6	easy · avg 78%	▇▇▇ 7.7%	typical
MultiPL-E lab	79.6% ✎	▄·▄····▄▄·▄█ beats 6/6	easy · avg 71%	▇▇▇ 7.0%	typical
MBPP+ lab	72.8% ✎	▄····▄▄·▄·█▄ beats 5/6	easy · avg 67%	▇▇▇ 6.5%	at its level
CodeReviewQA: change localisation (hard), invariant accuracy	37.3% ✎	█▂▁▁▂▂▂▃▅▅▅▇ beats 30/47	medium · avg 36%	▇▇ 4.9%	sharp · at its level · noisy
CRUXEval-I lab	69.5% ✎	█·██·█··█·██ beats 5/6	medium · avg 57%	▇ 3.6%	at its level

4 more tests

test	result	vs the others in its run	difficulty	counts	why
CodeReviewQA: solution identification (hard), invariant accuracy	56.2% ✎	▃··▁▂▂▃▂█▃▄▂ beats 31/47	medium · avg 54%	▇ 3.3%	sharp · at its level · noisy
CodeReviewQA: automated code refinement, exact match rate	46.6% ✎	▁▁▁▁▁▁▂▂▃█▅▄ beats 45/47	hard · avg 33%	▇ 3.2%	at its level
CRUXEval-O lab	79.5% ✎	▄·█··▄·▄··▄▄ beats 5/6	medium · avg 55%	▇ 1.3%	blunt · at its level
CodeReviewQA: change type recognition, invariant accuracy	73.9% ✎	▁▁▁······▁▃█ beats 27/47	easy · avg 67%	▇ 0.1%	blunt · at its level · noisy

other tests3 not in a task · 16% of its score

test	result	vs the others in its run	difficulty	counts	why
DSEffi-Bench Pass (1,000 data-science tasks, greedy)	32.8% ✎	▅··▃▅·█·▃██▃ beats 3/15	hard · avg 24%	▇▇▇ 8.4%	at its level
BigCodeBench (Instruct) lab	48.4% ✎	▄······▄▄█·█ beats 5/6	medium · avg 38%	▇▇ 5.1%	at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)	73.1% ✎	▃▁▁▁▁▁▁▁▂█▅▁ beats 204/276	medium · avg 73%	▇ 2.6%	at its level · noisy

$ where to run it

llama.cpp · recommended

version

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 ↗
Run Qwen2.5 Coder 14B Instruct · needs ~9 GB · from huggingface.co/lmstudio-community
chat in the terminal
llama cli -hf lmstudio-community/Qwen2.5-Coder-14B-Instruct-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf lmstudio-community/Qwen2.5-Coder-14B-Instruct-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Qwen2.5 Coder 14B Instruct · needs ~9 GB · from huggingface.co/lmstudio-community
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/lmstudio-community/Qwen2.5-Coder-14B-Instruct-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Qwen2.5 Coder 14B Instruct · needs ~9.36 GB · from ollama.com
chat in the terminal
ollama run qwen2.5-coder:14b-instruct-q4_K_M

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Qwen2.5 Coder 14B Instruct · needs ~8.3 GB, Mac only · from huggingface.co/lmstudio-community
chat in the terminal
mlx_lm.chat --model lmstudio-community/Qwen2.5-Coder-14B-Instruct-MLX-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model lmstudio-community/Qwen2.5-Coder-14B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · Qwen (Alibaba) vendor-reporte… 53%, Aider code editing benchmark… 23%, CodeReviewQA: The Code Review… 11% and 2 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

source	results	share of the score	licence
Qwen (Alibaba) vendor-reported benchmark scores for its own models (qwen.ai launch posts, Qwen2.5-VL and Qwen3-VL technical reports, Qwen3.8-Flash-Next card) · lab-reported	8	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 53%	Factual vendor-reported score (not licensed content)
Aider code editing benchmark runs (old edit leaderboard)	1	▇▇▇▇▇▇▇ 23%	Apache-2.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)	4	▇▇▇ 11%	CC-BY-4.0
DSEffi-Bench: Demystifying Large Language Models' Capability in Efficient Data Science Code Generation — authors' own model runs	1	▇▇▇ 8%	CC BY 4.0
Open Portuguese LLM Leaderboard (archived)	1	▇ 3%	Apache-2.0
prior · pull toward the middle		▇ 1%
weighted average		= 55.1

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-10-11 02:20 UTC · sources and licences · json