Contamination Map Methodology

Text / LLM / Model

InternVL3.5

Organisation
Shanghai AI Laboratory (OpenGVLab)
Family
InternVL · version 3.5
Released
2025-08-25 (arXiv v1 submission 2025-08-25 and HF repo creation of OpenGVLab/InternVL3_5-* (2025-08-25))
Checkpoints
OpenGVLab/InternVL3_5-1B, OpenGVLab/InternVL3_5-2B, OpenGVLab/InternVL3_5-4B, OpenGVLab/InternVL3_5-8B, OpenGVLab/InternVL3_5-14B, OpenGVLab/InternVL3_5-38B, OpenGVLab/InternVL3_5-30B-A3B, OpenGVLab/InternVL3_5-241B-A28B, OpenGVLab/InternVL3_5-GPT-OSS-20B-A4B-Preview, OpenGVLab/InternVL3_5-8B-Pretrained, OpenGVLab/InternVL3_5-8B-Instruct, OpenGVLab/InternVL3_5-8B-MPO, OpenGVLab/InternVL3_5-8B-Flash
Weights
Open · Apache 2.0
Links
PaperModel cardCode

Training data disclosure: Partial list

Level determined for 18% of benchmark cells.

“Multimodal data: this subset of data is mainly sourced from the training corpora of InternVL3 [187], covering a diverse range of domains such as image captioning, general question answering, mathematics, scientific disciplines, charts, optical character recognition (OCR), knowledge grounding, document understanding, multi-turn dialogue, and medical data.”
arxiv.org · paper · Section 2.2 Data · retrieved 2026-10-15
“Instruction-following data from InternVL3, which are reused to preserve broad coverage of vision–language tasks.”
arxiv.org · paper · Section 2.3 Supervised Fine-Tuning · retrieved 2026-10-15
“We originally intended to open-source the training dataset used in InternVL. However, due to the company policy, we have withdrawn the dataset from public access.”
huggingface.co · dataset card · Note on Dataset Availability · retrieved 2026-10-15

Built on

Everything these base models may have seen is inherited by InternVL3.5.

Other declared training data

Benchmark cells

Classic LLM benchmarks

Unknown MMLU (Measuring Massive Multitask Language Understanding) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown GSM8K (Grade School Math 8K) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown MATH (Mathematics Aptitude Test of Heuristics) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown HumanEval · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown ARC-Challenge (AI2 Reasoning Challenge, Challenge Set) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown HellaSwag · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown TruthfulQA · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown WinoGrande · unknown for its base model Qwen3: training data not disclosed in enough detail

Multimodal LLM benchmarks

Unknown MMMU (Massive Multi-discipline Multimodal Understanding and Reasoning) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown MMMU-Pro · unknown for its base model Qwen3: training data not disclosed in enough detail

Related data seen MathVista · in fine-tuning data: related dataset ChartQA was used

Evidence (6)
“Instruction-following data from InternVL3, which are reused to preserve broad coverage of vision–language tasks.”
arxiv.org · paper · Section 2.3 Supervised Fine-Tuning · retrieved 2026-10-15
“For SFT data, we construct the training corpora based on those used in InternVL2.5 [18] while introducing additional tool usage, 3D scene understanding, GUI operations, scientific diagrams, creative writing, and multimodal reasoning samples.”
arxiv.org · paper · Section 2.3 Data · retrieved 2026-10-15
“ChartQA (en) [181], MMTab (en) [310], PlotQA (en) [187], FigureQA (en) [105], VisText (en) [226]”
arxiv.org · paper · Table 5 (InternVL 2.5 fine-tuning data mixture), Chart row · retrieved 2026-10-15
“It consists of 6,141 examples, derived from 28 existing multimodal datasets involving mathematics and 3 newly created datasets”
arxiv.org · paper · Abstract · retrieved 2026-10-15
“ChartQA (2022)”
arxiv.org · paper · Table 5 (Appendix C) · retrieved 2026-10-15
“"dataset": "ChartQA",”
huggingface.co · dataset card · source.json · retrieved 2026-10-15

Seen, split unclear ChartQA · in fine-tuning data: the dataset was used; the source does not say which part

Evidence (3)
“Instruction-following data from InternVL3, which are reused to preserve broad coverage of vision–language tasks.”
arxiv.org · paper · Section 2.3 Supervised Fine-Tuning · retrieved 2026-10-15
“For SFT data, we construct the training corpora based on those used in InternVL2.5 [18] while introducing additional tool usage, 3D scene understanding, GUI operations, scientific diagrams, creative writing, and multimodal reasoning samples.”
arxiv.org · paper · Section 2.3 Data · retrieved 2026-10-15
“ChartQA (en) [181], MMTab (en) [310], PlotQA (en) [187], FigureQA (en) [105], VisText (en) [226]”
arxiv.org · paper · Table 5 (InternVL 2.5 fine-tuning data mixture), Chart row · retrieved 2026-10-15

Seen, split unclear DocVQA · in fine-tuning data: the dataset was used; the source does not say which part

Evidence (3)
“Instruction-following data from InternVL3, which are reused to preserve broad coverage of vision–language tasks.”
arxiv.org · paper · Section 2.3 Supervised Fine-Tuning · retrieved 2026-10-15
“For SFT data, we construct the training corpora based on those used in InternVL2.5 [18] while introducing additional tool usage, 3D scene understanding, GUI operations, scientific diagrams, creative writing, and multimodal reasoning samples.”
arxiv.org · paper · Section 2.3 Data · retrieved 2026-10-15
“DocVQA (en) [42], Docmatix (en) [121], DocReason25K (en) [93], Sujet-Finance-QA-Vision (en) [217]”
arxiv.org · paper · Table 5 (InternVL 2.5 fine-tuning data mixture), Document row · retrieved 2026-10-15

Seen, split unclear AI2D (AI2 Diagrams) · in fine-tuning data: the dataset was used; the source does not say which part

Evidence (3)
“Instruction-following data from InternVL3, which are reused to preserve broad coverage of vision–language tasks.”
arxiv.org · paper · Section 2.3 Supervised Fine-Tuning · retrieved 2026-10-15
“For SFT data, we construct the training corpora based on those used in InternVL2.5 [18] while introducing additional tool usage, 3D scene understanding, GUI operations, scientific diagrams, creative writing, and multimodal reasoning samples.”
arxiv.org · paper · Section 2.3 Data · retrieved 2026-10-15
“AI2D (en) [109], ScienceQA (en) [165], TQA (en) [110], ChemVLM Data (en & zh) [129]”
arxiv.org · paper · Table 5 (InternVL 2.5 fine-tuning data mixture), Science row · retrieved 2026-10-15

Unknown CharXiv · unknown for its base model Qwen3: training data not disclosed in enough detail

Recent LLM benchmarks

Unknown GPQA Diamond · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown MMLU-Pro · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown LiveCodeBench (code generation, release_v6) · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown SWE-bench Verified · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown AIME 2024 · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown AIME 2025 · unknown for its base model Qwen3: training data not disclosed in enough detail

Unknown MATH-500 · unknown for its base model Qwen3: training data not disclosed in enough detail

Notes

Latest InternVL release found (no InternVL4 as of 2026-10-15). Benchmark-named data only reachable through the InternVL3 -> InternVL2.5 chain; the SFT data release was withdrawn. InternVL3.5-Flash variants add a router trained on a subset of the SFT data. No decontamination statement found. Accepts image input: “We initialize the language model using the Qwen3 series and GPT-OSS, and the vision encoder using InternViT-300M and InternViT-6B.”

Sources