whatmodel.sh / GLM-4 9B Chat

GLM-4 9B Chat

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

GLM-4 9B Chat is an open-weights language model from Zhipu AI (Z.ai), released June 2024. Its published size is 9.4B parameters and a 128k-token context window. Licence: glm-4.

Catalogue identity notes

Maker repo zai-org/glm-4-9b-chat @bd8234f: licence tag other (glm-4). 9,399,951,392 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-06: month the maker's Hugging Face repo was created (createdAt 2024-06-04). The repo was previously published as THUDM/glm-4-9b-chat (the Hugging Face API redirects that id here); leaderboard rows may use the old id. Ollama tags glm4:9b-chat-q4_K_M, glm4:9b-chat-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.

47.2±1.8#395 of 719 LLMsfrom 8 benchmark results across 7 sources

$ how it compares

The LLMs ranked either side of GLM-4 9B Chat, out of 828 with a score.

#modelscorewhereRAMreleased
470Gemma 2 9B IT Google47±1local~7GB2024-06
471Command R Plus (08-2024) Cohere47±1api $4.38/M +local~63GB2024-08
472GLM-4 9B Chat Zhipu AI (Z.ai)47±2local~7.9GB2024-06
473Yi-1.5 34B (base) 01.AI47±3local~22GB2024-05
474Phi-4-mini-reasoning 3.8B Microsoft47±2api $0.13/M +local~3.8GB2025-04

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #284 of 300 · 20% of its score · 3 tests

Trails Granite 4.0 H 350M, ChatGLM3-6B, Qwen3 0.6B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL V4 (2025-08-26 archive, models absent from later archives)21.2% ✎▄▄····█▄···▄ beats 1/5hard · avg 27%▇▇▇ 7.8%typical
BFCL v339.1% ✎▁▁▁▁▃▃▅█▅▄▁▁ beats 12/114medium · avg 52%▇▇▇ 6.5%blunt · at its level
BFCL (version not stated) lab81.0% ✎█·········██ beats 2/5easy · avg 78%▇▇ 5.6%blunt
other tests5 not in a task · 79% of its score
testresultvs the others in its rundifficultycountswhy
WildBench v2 WB-Score (task macro)39.0990 WB-Score ✎▁·▁▁·▃█▄▇▆▅▆ beats 28/52avg 45.199 WB-Score▇▇▇▇▇▇▇▇▇▇▇▇▇ 33%sharp
CPTUB average (SpeakLeash, 0-5 judge scale)59.0% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 28/95easy · avg 66%▇▇▇▇▇▇▇▇ 21%at its level
Polish EQ-Bench v2 (SpeakLeash)61.8% ✎▁··▁▁▂▂▆▅▆█▆ beats 43/70medium · avg 64%▇▇▇▇▇▇▇ 17%at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)75.0% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 225/276medium · avg 73%▇▇▇ 6.4%at its level · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)98.7% ✎▁▁▂▃▄▄█▆▃▂▃▁ beats 133/140very easy · avg 94%▇ 2.0%blunt · maxed out · 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 GLM-4 9B Chat · needs ~6.3 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/glm-4-9b-chat-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/glm-4-9b-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 GLM-4 9B Chat · needs ~6.3 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/glm-4-9b-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 GLM-4 9B Chat · needs ~6.14 GB · from ollama.com
    chat in the terminal
    ollama run glm4:9b-chat-q4_K_M

Weights and documentation: huggingface.co/zai-org ↗

$ how the score was worked out · WildBench v2 leaderboard data… 33%, SpeakLeash CPTUB (Complex Pol… 21%, SpeakLeash Polish EQ-Bench le… 17% and 4 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
WildBench v2 leaderboard data (Ai2)1▇▇▇▇▇▇▇▇▇▇ 33%Apache-2.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇▇▇▇▇▇ 21%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇▇▇▇▇ 17%Apache-2.0
Berkeley Function Calling Leaderboard result archive2▇▇▇▇ 14%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇▇ 6%Apache-2.0
ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools (Team GLM, arXiv 2406.12793v2) · lab-reported1▇▇ 6%CC BY 4.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 2%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 47.2

$ 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