[whatmodel.sh](../../README.md) / CogVLM2 Llama3 Chat 19B # CogVLM2 Llama3 Chat 19B Zhipu AI (THUDM) · released 2024-05 · 19.5B params · cogvlm2 **score 44.2** ±2.7 · #491 of 732 · from 5 benchmark results across 3 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks CogVLM2 Llama3 Chat 19B among every model on that task or test, worst to best.}

task                vs the others         rank  counts  tests
vision & documents  ▁●▂▂▁▃▄▂▆▇█▄   #146 of 153     97%      5

vision & documents · #146 of 153 · 97% of its score

Beats [Gemma 3 4B](gemma-3-4b.md) here, though they rank higher overall. Trails [Qwen3.5 0.8B](qwen3-5-0-8b.md), [North Micro Vision Instruct (2.4B)](north-micro-vision-instruct.md), [Molmo 7B-D (0924)](molmo-7b-d.md) here, though they rank lower overall.

test                                 result  vs the others in its run  difficulty      counts  why
Arena vision (style controlled)       963.7  ▁▁●▂▂▂▃▄▅▅█▇ beats 4/152  avg 1061           51%  blunt
CapArena-Auto detailed image capt…  -8.5000  ▃▃▃·▃●·▅▃▃▃▃ beats 6/12   avg -4.498         32%  typical
ChartQA (maker)                       81.0%  ▁··▁▁▁▁▁▂●▅█ beats 15/44  easy · avg 80%    6.4%  blunt · small run
OCRBench (maker)                      75.6%  ▁·▃●▆▅▂█▅█▁▃ beats 4/37   easy · avg 82%    5.1%  blunt · small run
TextVQA (maker)                       84.2%  ▁▁▁▄▅▁▃·▃▃▃● beats 19/24  easy · avg 78%    2.9%  blunt · 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                     1   51% ▇▇▇▇▇▇
CapArena: Benchmarking and Analyzing De…        1   32% ▇▇▇▇
Zhipu AI (submitted / maker-reported)           3   14% ▇▇
prior: pull toward the middle                        3% ▇
weighted average                                   = 44.2

_{Licences: LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/))} ## How to run it ### Other ways to run it 1. Run it as published (BF16, needs ~39.01 GB, NVIDIA only, from [github.com/zai-org](https://github.com/zai-org/CogVLM2/blob/6700ae7ca23ba87f13796f3f427cbda64f71aaef/basic_demo/README.md)) ```sh git clone https://github.com/zai-org/CogVLM2 && cd CogVLM2/basic_demo && pip install -r requirements.txt && CUDA_VISIBLE_DEVICES=0 python cli_demo.py ``` Other versions: int4 (bitsandbytes at load) (~39.01 GB, NVIDIA only) `git clone https://github.com/zai-org/CogVLM2 && cd CogVLM2/basic_demo && pip install -r requirements.txt && CUDA_VISIBLE_DEVICES=0 python cli_demo.py --quant 4` Model repository and documentation: [huggingface.co/zai-org](https://huggingface.co/zai-org/cogvlm2-llama3-chat-19B) [interactive version](https://whatmodel.sh/model/cogvlm2-llama3-chat-19b) · [JSON](https://whatmodel.sh/api?format=json&all=1)