InternVL3 5 8B — Hardware Requirements & GPU Compatibility
Vision
InternVL3.5-8B is OpenGVLab's 8.5-billion-parameter multimodal model, combining a 0.3-billion-parameter vision encoder with an 8.2-billion-parameter language model. The card says the InternVL3.5 family uses a Cascade Reinforcement Learning framework, offline RL followed by online RL, to improve reasoning and versatility across multimodal, text and agentic tasks, and this checkpoint is the finetune of InternVL3_5-8B-MPO. It is suited to image and document question answering, reasoning over visual content and agent-style tasks. At this size it runs on a single consumer GPU once quantized. The card does not state a context length, so none is given here. It is released under the Apache 2.0 license, permitting commercial and research use. Published in August 2025, it sits in the middle of the InternVL3.5 line, which spans 1B to 38B dense models plus larger mixture-of-experts variants up to 241B total parameters.
44.6K downloads
108 likesSpecifications
- Publisher
- OpenGVLab
- Parameters
- 8.5B
- Architecture
- InternVLChatModel
- Release Date
- 2025-08-25
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does InternVL3 5 8B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 4.0 GB | — | 3.62 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 4.1 GB | — | 3.73 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 4.6 GB | — | 4.16 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 4.7 GB | — | 4.26 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 5.6 GB | — | 5.12 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 6.7 GB | — | 6.08 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 7.7 GB | — | 7.04 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 9.4 GB | — | 8.53 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run InternVL3 5 8B?
Q4_K_M · 5.6 GBInternVL3 5 8B (Q4_K_M) requires 5.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 8+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run InternVL3 5 8B?
Q4_K_M · 5.6 GB58 devices with unified memory can run InternVL3 5 8B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroomWhere to Download InternVL3 5 8B
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does InternVL3 5 8B need?
InternVL3 5 8B requires 5.6 GB of VRAM at Q4_K_M, or 18.8 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 8.5B × 4.8 bits ÷ 8 = 5.1 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
Q4_K_M5.6 GB- What's the best quantization for InternVL3 5 8B?
For InternVL3 5 8B, Q4_K_M (5.6 GB) offers the best balance of quality and VRAM usage. Q4_K_L (5.8 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 2.6 GB.
VRAM requirement by quantization
IQ2_XXS2.6 GBIQ3_XS3.9 GBQ3_K_L4.8 GBQ4_K_M ★5.6 GBQ4_K_L5.8 GBBF1618.8 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run InternVL3 5 8B on a Mac?
InternVL3 5 8B requires at least 2.6 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run InternVL3 5 8B locally?
Yes — InternVL3 5 8B can run locally on consumer hardware. At Q4_K_M quantization it needs 5.6 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is InternVL3 5 8B?
At Q4_K_M, InternVL3 5 8B can reach ~853 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~116 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 5.6 × 0.65 = ~924 tok/s
Estimated speed at Q4_K_M (5.6 GB)
~924 tok/s~116 tok/s~924 tok/s~853 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of InternVL3 5 8B?
At Q4_K_M, the download is about 5.12 GB. The full-precision BF16 version is 17.06 GB. The smallest option (IQ2_XXS) is 2.35 GB.
- Which GPUs can run InternVL3 5 8B?
52 consumer GPUs can run InternVL3 5 8B at Q4_K_M (5.6 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 40 GPUs have plenty of headroom for comfortable inference.
- Which devices can run InternVL3 5 8B?
59 devices with unified memory can run InternVL3 5 8B at Q4_K_M (5.6 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.