InternVL3 5 8B — Hardware Requirements & GPU Compatibility
VisionInternVL3.5-8B is OpenGVLab's 8.5-billion-parameter multimodal model, which pairs a 0.3-billion-parameter vision encoder with an 8.2-billion-parameter language model. The card notes that the InternVL3.5 family uses Cascade Reinforcement Learning, applying offline RL before online RL to boost reasoning and versatility across multimodal, text, and agentic tasks; this checkpoint is a finetune of InternVL3_5-8B-MPO. It works well for image and document question answering, reasoning over visual content, and agent-style tasks. At this size it can run on a single consumer GPU once quantized. The card doesn't state a context length, so none is provided here. Released under the Apache 2.0 license, it allows both commercial and research use. Published in August 2025, it falls in the middle of the InternVL3.5 lineup, which includes dense models ranging from 1B to 38B parameters plus larger mixture-of-experts variants reaching up to 241B total parameters.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 8K)Best on 24 GB (Q4_K_M · ~116 tok/s · 32K)Macs from 16 GB (Q4_K_M · ~25 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run InternVL3 5 8B
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/bartowski/OpenGVLab_InternVL3_5-8B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 15.8 GB free · 16K context
Download: OpenGVLab_InternVL3_5-8B-Q4_K_M.gguf5.03 GBbartowski
Ways to runOther ways
each opens step-by-stepRun in cloud
Your own private copy · Q4_K_M · 24 GB GPU
- The easiest way to try this model
- Running in a few minutes, easy setup
- Chat right here or connect any app
$5+ bonus for new users
Runs on RTX 3080 (10 GB) (1 GB headroom) · Q4_K_M
- Generation speed
- ~88 tok/s
- generation speed
- Cost per 1M output tokens
- $0.38
- per 1M output tokens
How Much VRAM Does InternVL3 5 8B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 4.0 GB | 9.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~164 tok/s | |
| 4.1 GB | 9.7 GB | 3-bit small — 3-bit small quantization | Fits: 32K context · ~160 tok/s | |
| 4.6 GB | 10.2 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~143 tok/s | |
| 4.7 GB | 10.3 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 32K context · ~140 tok/s | |
| 5.6 GB | 11.2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~116 tok/s | |
| 6.7 GB | 12.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~98 tok/s | |
| 7.7 GB | 13.3 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~85 tok/s | |
| 9.4 GB | 15 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~70 tok/s |
Selected Q4_K_M
4-bit medium quantization8.2GBat 16K context
- Weights
- 5.0 GB
- KV cache
- 3.0 GB
- Overhead
- 0.3 GB
Which Hardware 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. 107 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti. 172 devices with unified memory can run InternVL3 5 8B, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB), NVIDIA Jetson Orin Nano 8GB (Super).
12 GB
12 GB tier →fits · 3.8 GB free · up to 32K ctx
16 GPUs · 1 device family
RTX 3080 12GB~105Simulate output on RTX 3080 12GBRTX 3080 Ti~105Simulate output on RTX 3080 TiRTX 5070~78Simulate output on RTX 5070iPhone~10Simulate output on Apple iPhone 17 Pro20 GB
fits · 11.8 GB free · up to 32K ctx
1 GPU
RX 7900 XT~85Simulate output on RX 7900 XT32–40 GB
fits · 23.8 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 9 device families
RTX 5090~207Simulate output on RTX 5090M5 Max*36 GB~57Simulate output on Mac Studio M5 Max (36 GB)M632 GB~21Simulate output on iMac 24" M6 (32 GB)RTX Spark N1X~35Simulate output on NVIDIA RTX Spark N1X Laptop (32 GB)Orin~24Simulate output on NVIDIA Jetson AGX Orin 32GB
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite Switch 4.1 8BIBM | — | 8.4 GB | ~111 | same size |
| Apertus V1.5 8Bswiss-ai | — | 8.5 GB | ~111 | same size |
| LFM2.5 8B A1BLiquid AI | — | 5.6 GB | ~335 | same size |
| BitCPM CANN 8BOpenBMB | — | 5.6 GB | ~127 | same size |
| Gemma 4 E4BGoogle | — | 5.6 GB | ~123 | same size |
| Qwen3 VL 8B InstructAlibaba | — | 8 GB | ~112 | same size |
Specifications
- Publisher
- OpenGVLab
- Parameters
- 8.5B
- Architecture
- InternVLChatModel
- Release Date
- 2025-08-25
- License
- Apache 2.0
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)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
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
★ Recommended — best balance of quality and VRAM usage.
Can I run InternVL3 5 8B on a Mac?
Yes — Mac Mini M2 Pro (16 GB) and 145 other Macs can run InternVL3 5 8B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 5.6 GB of usable unified memory (RAM minus macOS overhead).
Can I run InternVL3 5 8B locally?
Yes — InternVL3 5 8B can run locally on consumer hardware. 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 ÷ 5.63 × 0.65 = ~924 tok/s
Estimated speed at Q4_K_M (5.6 GB)
Real-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?
107 consumer GPUs can run InternVL3 5 8B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6700M, AMD Radeon RX 6750 XT, AMD Radeon RX 5600 XT. 51 GPUs have plenty of headroom for comfortable inference.
Which devices can run InternVL3 5 8B?
172 devices with unified memory can run InternVL3 5 8B at Q4_K_M (5.6 GB), including AMD Ryzen 7 8845HS (Hawk Point) Laptop, AMD Ryzen AI 7 350 (Krackan Point) Laptop, AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.