InternVL3 5 1B — Hardware Requirements & GPU Compatibility
VisionInternVL3.5-1B is OpenGVLab's 1.1-billion-parameter multimodal model, combining a 0.3-billion-parameter vision encoder with a 0.8-billion-parameter language model. As the smallest member of the InternVL3.5 family, it uses a Cascade Reinforcement Learning framework that applies offline RL before online RL to boost reasoning capabilities; this specific version was fine-tuned from the InternVL3_5-1B-MPO checkpoint. The card covers general multimodal tasks as well as reasoning, text, and agentic benchmarks. Its size makes it easy enough to run on nearly any consumer GPU, a laptop, or even a CPU after quantization. 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 represents the smallest entry in a series ranging from 1B up to a 241B mixture-of-experts model, with dense siblings of 2B, 4B, 8B, 14B and 38B models in between.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~936 tok/s · 32K)Macs from 8 GB (Q4_K_M · ~102 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run InternVL3 5 1B
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_5-1B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 21.8 GB free · 16K context
Download: OpenGVLab_InternVL3_5-1B-Q4_K_M.gguf0.48 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 3060 12GB (9 GB headroom) · Q4_K_M
- Generation speed
- ~334 tok/s
- generation speed
- Cost per 1M output tokens
- $0.052
- per 1M output tokens
How Much VRAM Does InternVL3 5 1B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.5 GB | 3.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~1310 tok/s | |
| 0.5 GB | 3.6 GB | 3-bit small — 3-bit small quantization | Fits: 32K context · ~1285 tok/s | |
| 0.6 GB | 3.7 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~1150 tok/s | |
| 0.6 GB | 3.7 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 32K context · ~1130 tok/s | |
| 0.7 GB | 3.8 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~936 tok/s | |
| 0.8 GB | 4 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~789 tok/s | |
| 1.0 GB | 4.1 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~683 tok/s | |
| 1.2 GB | 4.3 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~560 tok/s |
Selected Q4_K_M
4-bit medium quantization2.2GBat 16K context
- Weights
- 0.2 GB
- KV cache
- 1.7 GB
- Overhead
- 0.3 GB
Which Hardware Can Run InternVL3 5 1B?
Q4_K_M · 0.7 GBInternVL3 5 1B (Q4_K_M) requires 0.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. 114 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 185 devices with unified memory can run InternVL3 5 1B, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB).
≤ 11 GB
≤ 11 GB tier →fits · 1.8 GB free · up to 32K ctx
68 GPUs · 1 Mac class · 3 device families
RTX 3080~706Simulate output on RTX 3080RTX 2080 Ti~572Simulate output on RTX 2080 TiRTX 3070 Ti~565Simulate output on RTX 3070 TiM3*8 GB~102Simulate output on iMac 24" M3 (8 GB)Orin Nano~95Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~68Simulate output on iPhone 1720 GB
fits · 17.8 GB free · up to 32K ctx
1 GPU
RX 7900 XT~686Simulate output on RX 7900 XT32–40 GB
fits · 29.8 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 9 device families
RTX 5090~1664Simulate output on RTX 5090M5 Max*36 GB~460Simulate output on Mac Studio M5 Max (36 GB)M632 GB~170Simulate output on iMac 24" M6 (32 GB)RTX Spark N1X~279Simulate output on NVIDIA RTX Spark N1X Laptop (32 GB)Orin~190Simulate 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 |
|---|---|---|---|---|
| MiniCPM5 1BOpenBMB | — | 1.3 GB | ~662 | same size |
| Gemma 3 1B ITGoogle | — | 2.1 GB | ~993 | same size |
| TinyLlama 1.1B Chat v1.0TinyLlama | — | 1 GB | ~649 | same size |
| LLaMA3.1 8B Instruct DFlash UltraChatz-lab | — | 2.7 GB | ~269 | same size |
| Vaultgemma 1BGoogle | — | 3.7 GB | ~286 | same size |
| Qwen3 8B Speculator.eagle3RedHatAI | — | 2.1 GB | ~978 | same size |
Specifications
- Publisher
- OpenGVLab
- Parameters
- 1.1B
- Architecture
- InternVLChatModel
- Release Date
- 2025-08-25
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does InternVL3 5 1B need?
InternVL3 5 1B requires 0.7 GB of VRAM at Q4_K_M, or 2.3 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 1.1B × 4.8 bits ÷ 8 = 0.6 GB
KV Cache + Overhead ≈ 0.1 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 1B?
For InternVL3 5 1B, Q4_K_M (0.7 GB) offers the best balance of quality and VRAM usage. Q4_K_L (0.7 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 0.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run InternVL3 5 1B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 158 other Macs can run InternVL3 5 1B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.7 GB of usable unified memory (RAM minus macOS overhead).
Can I run InternVL3 5 1B locally?
Yes — InternVL3 5 1B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is InternVL3 5 1B?
At Q4_K_M, InternVL3 5 1B can reach ~6857 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~936 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 ÷ 0.7 × 0.65 = ~7429 tok/s
Estimated speed at Q4_K_M (0.7 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 1B?
At Q4_K_M, the download is about 0.64 GB. The full-precision BF16 version is 2.12 GB. The smallest option (IQ2_XXS) is 0.29 GB.
Which GPUs can run InternVL3 5 1B?
114 consumer GPUs can run InternVL3 5 1B at Q4_K_M. Top options include AMD Radeon RX 5600 XT, AMD Radeon RX 5700 XT, AMD Radeon RX 580 8GB. 114 GPUs have plenty of headroom for comfortable inference.
Which devices can run InternVL3 5 1B?
185 devices with unified memory can run InternVL3 5 1B at Q4_K_M (0.7 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.