Skip to content
Local AI Wizard
Your machine
Theme

LLaVA-OneVision-1.5-8B-Instruct

LMMs-Lab · lmms-lab/LLaVA-OneVision-1.5-8B-Instruct

Weights license: Apache 2.0 · code and components may use other terms

Reported by the publisher27 resultsLMMs-Lab's model card · read 2026-10-25 · not ranked · see them
Memory needed~7 GBQ4 weights + working room, before the machine's reserve
Parameters8.5 BHub safetensors total
Repository updated2026-07-16A commit is not a release date

Quality

No comparable score

We track no public, comparable benchmark that covers this model yet. No score is inferred.

Reported by the publisher

27 results · LMMs-Lab's model card · read 2026-10-25
MMMU (Validation)
55.44lmms_eval
MMMU-Pro (Standard)
37.40lmms_eval
MMMU-Pro (Vision)
25.15lmms_eval
MMBench (English; Test)
84.14lmms_eval
MMBench (Chinese; Test)
81.00lmms_eval
MME-RealWorld (English)
62.31lmms_eval
MME-RealWorld (Chinese)
56.11lmms_eval
AI2D (With Mask)
84.16lmms_eval
Every reported result27
AI2D (Without Mask)
94.11lmms_eval
CV-Bench
80.82lmms_eval
VL-RewardBench
45.90lmms_eval
V
78.01lmms_eval
PixmoCount
62.19lmms_eval
CountBench
88.19lmms_eval
ChartQA
86.48lmms_eval
CharXiv (Direct Questions)
74.10lmms_eval
DocVQA (Test)
95.00lmms_eval
InfoVQA (Test)
78.42lmms_eval
WeMath
33.62lmms_eval
MathVista (Mini)
69.57lmms_eval
MathVision
25.56lmms_eval
MMStar
67.72lmms_eval
SEED-Bench (Image)
77.32lmms_eval
ScienceQA
94.98lmms_eval
SEED-Bench 2-Plus
69.21lmms_eval
OCRBench
82.90lmms_eval
RealWorldQA
68.10lmms_eval

Values as printed on the card, with the publisher's own settings (card section: Evaluation Results); column “LLaVA-OV-1.5-8B”. Not comparable across publishers and not used in any ranking. Model card

Runs on

Fast memory holds it 90

The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.

Show all 90

Within memory, outside the fast part 26

Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.

Show all 26

One memory tier short 3

The same machines with the next memory size up would hold it in fast memory.

Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Not a measured fit or speed.

Consider it for

LMMs-Lab released LLaVA-OneVision-1.5-8B-Instruct on 2025-09-28: a vision-language model that answers about images, with 8.53B parameters on the Hugging Face count and the Apache 2.0 license.

Check first

License: commercial use allowed. At Q4 the weights load about 6.4 GB before each machine's reserve; image resolution, video frames and context add memory, and the vision encoder needs support in the runtime you pick.

Measured speed

No published measurement for this checkpoint

Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.

How to run

Runtime compatibility and tokens per second have not been measured for this entry. Published metadata and a memory estimate do not prove a successful local run.

Library in source: transformers

Documented ways to run it

Copied as printed in the linked source (read 2026-10-18); not run by this site.

Transformers · model card · Using 🤗 Transformers to Chat
from transformers import AutoTokenizer, AutoProcessor, AutoModelForCausalLM
from qwen_vl_utils import process_vision_info
model_path = "lmms-lab/LLaVA-OneVision-1.5-8B-Instruct"

# default: Load the model on the available device(s)
model = AutoModelForCausalLM.from_pretrained(
    model_path, torch_dtype="auto", device_map="auto", trust_remote_code=True
)

# default processer
processor = AutoProcessor.from_pretrained(model_path, trust_remote_code=True)

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg",
            },
            {"type": "text", "text": "Describe this image."},
        ],
    }
]

# Preparation for inference
text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
    text=[text],
    images=image_inputs,
    videos=video_inputs,
    padding=True,
    return_tensors="pt",
)
inputs = inputs.to("cuda")

# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=1024)
generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text)

Choose my computer

Technical details

What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.

Task
image-text-to-text
Architecture
LLaVAOneVision1_5_ForConditionalGeneration
Library in source
transformers
Weight license
Apache 2.0lmms-lab/LLaVA-OneVision-1.5-8B-Instruct model card (Hugging Face) · read 2026-10-18source
Commercial use
Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
Released
LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training (arXiv 2509.23661, LMMs-Lab) · read 2026-10-18source
Every stored field13 fields
Revision checked
e137910552d616eb0f7305147ca054311a90c542
Architecture
LLaVAOneVision1_5_ForConditionalGeneration
Library
transformers
Hub task tag
image-text-to-text
Declared base models
Qwen/Qwen3-8B-Base · DeepGlint-AI/rice-vit-large-patch14-560
Formats
transformers, safetensors
Access
Open download
Publisher kind
Official
Hub parameter count
8,527,214,624safetensors.total
Repository created
2025-09-16creation is not the model's release
Repository updated
2026-07-16a commit is not a release date
Hub downloads (30 days) · likes
6,372 · 68this repository only · read 2026-10-25
Catalog state
Hub config or weights found; no inventory benchmark link

Sources

Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.

Curated facts checked: 2026-10-18. The license and release date cite the publisher's sources; parameters, files and task come from the Hugging Face record, and memory is a labelled estimate from its parameter count. A value no source states stays empty; a repository commit is not a model release date.