[whatmodel.sh](../../README.md) / GLM-4.5V # GLM-4.5V Z.ai · released 2025-08 · 106B params (12B active) · mit **score 61.2** ±1.5 · #193 of 732 · from 12 benchmark results across 4 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks GLM-4.5V among every model on that task or test, worst to best.}

task                vs the others          rank  counts  tests
chat & writing      ▁▁▁▃▆▄▆█●▇▇▃   #163 of 402*     23%      1
vision & documents  ▁▁▂▂▁▃▄●▆▇█▄     #95 of 153     68%      8

chat & writing · #163 of 402 · 23% of its score

Beats [MiniMax-M2](minimax-m2.md), [MiniCPM5 2B](minicpm5-2b.md), [Granite 4.2 30B](granite-4-2-30b.md) here, though they rank higher overall. Trails [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md), [LFM2.5 2.6B](lfm2-5-2-6b.md) here, though they rank lower overall.

test            result  vs the others in its run    difficulty  counts  why
LMArena (text)  1352.1  ▁▁▁▂▃▃▃▆●▇█▃ beats 196/385  avg 1260       23%  blunt

vision & documents · #95 of 153 · 68% of its score

Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-5 nano](gpt-5-nano.md), [Apriel 1.6 15B Thinker](apriel-1-6-15b-thinker.md) here, though they rank higher overall. Trails [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md), [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Qwen3.5 4B](qwen3-5-4b.md) here, though they rank lower overall.

test                                result  vs the others in its run   difficulty        counts  why
CharXiv (reasoning questions)        63.0%  ▄·····▄▄·●·█ beats 3/5     hard · avg 31%       16%  at its level · several runs
Arena vision (style controlled)     1153.1  ▁▁▂▂▂▂▃●▅▅█▇ beats 49/152  avg 1061             15%  blunt
ChartArena chart parsing (English…   43.8%  ▁·▁·▃▃▁▇▃●·▄ beats 14/22   hard · avg 27%       12%  blunt · at its level
MMMU (val) (maker)                   75.4%  ▁····▁▁▃█▆●▅ beats 39/53   medium · avg 57%    6.5%  at its level · small run
MMMU Pro (maker)                     65.2%  ▂▃▁▃▆▇▃●▇██▅ beats 24/52   medium · avg 46%    6.2%  at its level · small run
CharXiv descriptive questions        90.8%  ▄····▄···●▄▄ beats 3/5     easy · avg 67%      5.1%  noisy
MMStar (maker)                       75.3%  ▃▅▂▃▂▂▆█●▆█▅ beats 21/35   medium · avg 62%    4.8%  at its level · small run
OCRBench (maker)                     86.5%  ▁·▃▁▆▅▂●▅█▁▃ beats 20/37   easy · avg 82%      1.5%  blunt · small run

other tests (2)

test                  result  vs the others in its run  difficulty        counts  why
LVBench (maker)        53.8%  ▇·●▅▅▄▄··▃▁▁ beats 6/28   hard · avg 48%      4.3%  at its level · small run
AndroidWorld (maker)   57.0%  ▁·▁▁▂▃●▅▂▂·▂ beats 10/24  medium · avg 40%    3.9%  at its level · small run

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's 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 is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
LMArena leaderboard dataset                     2   38% ▇▇▇▇▇
Z.ai GLM-V GitHub repository benchmark …        7   35% ▇▇▇▇
ServiceNow-AI/Apriel-1.6-15b-Thinker mo…        2   14% ▇▇
ChartArena: Benchmarking Chart Parsing …        1   12% ▇
prior: pull toward the middle                        1% ▇
weighted average                                   = 61.2

_{Licences: LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Z.ai GLM-V GitHub repository benchmark image ([Apache-2.0](https://github.com/zai-org/GLM-V/blob/9013a771cfe578d8a0c7dfc4548156f3a23e4890/LICENSE)) · ServiceNow-AI/Apriel-1.6-15b-Thinker model card ([MIT](../../licenses/cards-final-NOTICE.txt)) · ChartArena: Benchmarking Chart Parsing across Languages ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))} ## How to run it ### Through an API api (Z.ai) · $0.90/M blended · $0.6 in / $1.8 out per 1M tokens · [docs.z.ai](https://docs.z.ai/guides/vlm/glm-4.5v.md) _{[price source](https://docs.z.ai/guides/overview/pricing): USD per 1M tokens, Z.ai API pricing page, Vision Models table, GLM-4.5V: input $0.6, output $1.8; cached input $0.11 excluded. Currency: USD. Z.ai official pricing page, observed 2026-09-28. · [context source](https://huggingface.co/zai-org/GLM-4.5V/blob/ed47433b37111465ec527affaaddceff371bca04/README.md): Context length stated on the official model card.} ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run GLM-4.5V (Q4_K_M, needs ~64.63 GB, from [huggingface.co/ggml-org](https://huggingface.co/ggml-org/GLM-4.5V-GGUF/tree/401d1a7a3b6a28971d7b0884eb05c50445d8852e)) ```sh llama cli -hf ggml-org/GLM-4.5V-GGUF:Q4_K_M ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf ggml-org/GLM-4.5V-GGUF:Q4_K_M ``` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run GLM-4.5V (Q4_K_M, needs ~64.63 GB, from [huggingface.co/ggml-org](https://huggingface.co/ggml-org/GLM-4.5V-GGUF/tree/401d1a7a3b6a28971d7b0884eb05c50445d8852e)) ```sh lms get https://huggingface.co/ggml-org/GLM-4.5V-GGUF@Q4_K_M -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name Model repository and documentation: [huggingface.co/zai-org](https://huggingface.co/zai-org/GLM-4.5V) [interactive version](https://whatmodel.sh/model/glm-4-5v) · [JSON](https://whatmodel.sh/api?format=json&all=1)