mirror/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2026-05-02 08:09:24 -05:00
Code Issues Packages Projects Releases Wiki Activity

chore(model gallery): add opengvlab_internvl3_5-30b-a3b (#6143)

Browse Source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto
2025-08-26 16:39:46 +02:00
committed by GitHub
parent 4381e892b8
commit 1b3f66018b
1 changed files with 44 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
gallery/index.yaml
+44
View File
@@ -1,4 +1,48 @@
---
- &internvl35
name: "opengvlab_internvl3_5-30b-a3b"
url: "github:mudler/LocalAI/gallery/qwen3.yaml@master"
icon: https://cdn-uploads.huggingface.co/production/uploads/64006c09330a45b03605bba3/zJsd2hqd3EevgXo6fNgC-.png
urls:
- https://huggingface.co/OpenGVLab/InternVL3_5-30B-A3B
- https://huggingface.co/bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF
license: apache-2.0
tags:
- multimodal
- gguf
- GPU
- Cpu
- image-to-text
- text-to-text
description: |
We introduce InternVL3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series. A key innovation is the Cascade Reinforcement Learning (Cascade RL) framework, which enhances reasoning through a two-stage process: offline RL for stable convergence and online RL for refined alignment. This coarse-to-fine training strategy leads to substantial improvements on downstream reasoning tasks, e.g., MMMU and MathVista. To optimize efficiency, we propose a Visual Resolution Router (ViR) that dynamically adjusts the resolution of visual tokens without compromising performance. Coupled with ViR, our Decoupled Vision-Language Deployment (DvD) strategy separates the vision encoder and language model across different GPUs, effectively balancing computational load. These contributions collectively enable InternVL3.5 to achieve up to a +16.0% gain in overall reasoning performance and a 4.05 ×\times× inference speedup compared to its predecessor, i.e., InternVL3. In addition, InternVL3.5 supports novel capabilities such as GUI interaction and embodied agency. Notably, our largest model, i.e., InternVL3.5-241B-A28B, attains state-of-the-art results among open-source MLLMs across general multimodal, reasoning, text, and agentic tasks—narrowing the performance gap with leading commercial models like GPT-5. All models and code are publicly released.
overrides:
parameters:
model: OpenGVLab_InternVL3_5-30B-A3B-Q4_K_M.gguf
mmproj: mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
files:
- filename: OpenGVLab_InternVL3_5-30B-A3B-Q4_K_M.gguf
sha256: c352004ac811cf9aa198e11f698ebd5fd3c49b483cb31a2b081fb415dd8347c2
uri: huggingface://bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF/OpenGVLab_InternVL3_5-30B-A3B-Q4_K_M.gguf
- filename: mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
sha256: fa362a7396c3dddecf6f9a714144ed86207211d6c68ef39ea0d7dfe21b969b8d
uri: huggingface://bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF/mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
- !!merge <<: *internvl35
name: "opengvlab_internvl3_5-30b-a3b-q8_0"
urls:
- https://huggingface.co/OpenGVLab/InternVL3_5-30B-A3B
- https://huggingface.co/bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF
overrides:
parameters:
model: OpenGVLab_InternVL3_5-30B-A3B-Q8_0.gguf
mmproj: mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
files:
- filename: OpenGVLab_InternVL3_5-30B-A3B-Q8_0.gguf
sha256: 79ac13df1d3f784cd5702b2835ede749cdfd274f141d1e0df25581af2a2a6720
uri: huggingface://bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF/OpenGVLab_InternVL3_5-30B-A3B-Q8_0.gguf
- filename: mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
sha256: fa362a7396c3dddecf6f9a714144ed86207211d6c68ef39ea0d7dfe21b969b8d
uri: huggingface://bartowski/OpenGVLab_InternVL3_5-30B-A3B-GGUF/mmproj-OpenGVLab_InternVL3_5-30B-A3B-f16.gguf
- &lfm2
url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
name: "lfm2-vl-450m"