BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

InternVL 3.5 8B

By OpenGVLab · China

Updated 2026-07-13

vision chat
Parameters
8B
License
Apache 2.0
Context
32k
VRAM (Q4)
6 GB
Released
January 2025

Overview

OpenGVLab's 8B vision-language model leading MMMU among open models. Built at Shanghai AI Lab and released under Apache 2.0.

When to pick this model

  • Best-in-class 8B vision for OCR and chart understanding
  • Single-GPU multimodal deployments
  • Document and PDF analysis pipelines
  • Apache-licensed VLM for commercial products

VRAM requirements by quantization

VRAM REQUIRED (GB)812Q4_K_M6 GBQ5_K_M7 GBQ8_010 GBFP1616 GB
QuantizationVRAM required
Q4_K_M (recommended)6 GB
Q5_K_M7 GB
Q8_010 GB
FP16 (no quantization)16 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

In practice, InternVL 3.5 8B fits an 8 GB consumer card at Q4_K_M (6 GB). Stepping up to Q8_0 raises the footprint to 10 GB, and unquantized FP16 weights take 16 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.

Without a GPU, InternVL 3.5 8B needs roughly 10 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 10 tokens/sec on entry-level GPUs, on the order of 30 tokens/sec on a mid-range card, and up to 80 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.

What hardware do you need

The table below matches InternVL 3.5 8B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.

GPU memoryExample cardsBest fit for InternVL 3.5 8B
8 GBRTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GBQ5_K_M (7 GB used)
12 GBRTX 5070, RTX 5070 Ti Laptop, RTX 4080 LaptopQ8_0 (10 GB used)
16 GBRTX 5080, RTX 4080 Super, Radeon RX 9070 XTFP16 (16 GB used)
24 GBRTX 4090, Radeon RX 7900 XTX, RTX 5090 LaptopFP16 (16 GB used)
32 GBRTX 5090FP16 (16 GB used)

Which GPU should you buy to run InternVL 3.5 8B?

To run InternVL 3.5 8B locally at Q4, you need ~6 GB of VRAM. The best value for this is a RTX 5060 (8 GB VRAM).

Check RTX 5060 price on Amazon →

As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.

Published benchmark scores

BenchmarkScore
MMMU61.5
DocVQA94.1

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

To put InternVL 3.5 8B in context: its MMMU score of 61.5 ranks #5 of the 8 catalog models with a published MMMU result (catalog median 62.8). Scores are author-published and measured under different harnesses, so treat gaps of a few points as noise rather than a verdict.

Strengths

  • Top quality-per-parameter ratio in 8B vision
  • Strong OCR and chart understanding
  • Apache 2.0 license
  • Solid VQA and short-video performance

Limitations

  • 32k context limits long-document multimodal work
  • Weaker multilingual coverage than Qwen2-VL
  • No native long-context extension

Typical workloads

In our catalog grid, InternVL 3.5 8B is filed under Laptop Vision, MMMU — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: vision-language work — screenshots, charts, scanned documents.

The 32k-token context window covers long chats and mid-sized documents, though very large retrieval workloads will need chunking. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.

Architecture & training

Architecture: Dense vision · 8B · InternVL 3.5 · InternLM backbone

Training: OpenGVLab — OCR, VQA, charts, short videos, PDF documents.

Verdict

The benchmark-leading small open VLM for OCR and charts — the right pick when you need accuracy more than context length.

Quick start

ollama run internvl3.5:8b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Similar models worth comparing

Frequently asked questions

How much VRAM does InternVL 3.5 8B need?

At the recommended Q4_K_M quantization, InternVL 3.5 8B needs about 6 GB of VRAM. Q8_0 takes 10 GB, and unquantized FP16 weights take 16 GB.

Can InternVL 3.5 8B run without a GPU?

Yes — with roughly 10 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.

What context window does InternVL 3.5 8B support?

InternVL 3.5 8B supports a 32k-token context window (32,768 tokens).

Can I use InternVL 3.5 8B commercially?

Yes. InternVL 3.5 8B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.

How fast is InternVL 3.5 8B on consumer hardware?

Our compatibility engine estimates on the order of 30 tokens/sec on a mid-range GPU and up to 80 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.

Which quantization of InternVL 3.5 8B should I download first?

Start with Q4_K_M (6 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It already fits an 8 GB card at Q5_K_M.

Tools

Is InternVL 3.5 8B the right pick for you?

Compute self-hosted ROI → Back to catalog