InternVL2 1B — Hardware Requirements & GPU Compatibility
Vision
InternVL2-1B is OpenGVLab's smallest instruction-tuned vision-language model in the InternVL 2.0 family, combining an InternViT-300M-448px vision encoder with a Qwen2-0.5B-Instruct language model via an MLP projector. It handles document and chart comprehension, OCR, scene-text reading, and general visual question answering, and can take multiple images or video frames as input. At under a billion parameters, it is light enough to run on a CPU or a low-end consumer GPU. InternVL 2.0 models were trained with an 8K context window and improved multi-image and video handling over the earlier InternVL 1.5 generation. Context length is 8,192 tokens. It is released under the MIT license, permitting unrestricted commercial and research use, and was published in July 2024. It was later superseded by the InternVL2.5 series.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~1057 tok/s · 32K)Macs from 8 GB (Q4_K_M · ~116 tok/s)
MIT
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run InternVL2 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/mradermacher/InternVL2-1B-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 22.0 GB free · 16K context
Download: InternVL2-1B.Q4_K_M.gguf0.49 GBmradermacher
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (10 GB headroom) · Q4_K_M
- Generation speed
- ~957 tok/s
- generation speed
- Cost per 1M output tokens
- $0.035
- per 1M output tokens
How Much VRAM Does InternVL2 1B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.4 GB | 3.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~1489 tok/s | |
| 0.5 GB | 3.5 GB | 3-bit small — 3-bit small quantization | Fits: 32K context · ~1456 tok/s | |
| 0.5 GB | 3.5 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~1310 tok/s | |
| 0.5 GB | 3.5 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 32K context · ~1260 tok/s | |
| 0.6 GB | 3.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~1057 tok/s | |
| 0.7 GB | 3.8 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~885 tok/s | |
| 0.8 GB | 3.9 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~771 tok/s | |
| 1.0 GB | 4.1 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~636 tok/s |
Selected Q4_K_M
4-bit medium quantization2.0GBat 16K context
- Weights
- 0.1 GB
- KV cache
- 1.6 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 22.0 GB free · ~1057 tok/s
≥50 tok/s needs ≥48 GB/s · RTX 5090 ~1879, RTX 3090 Ti ~1057
Which Hardware Can Run InternVL2 1B?
Q4_K_M · 0.6 GBInternVL2 1B (Q4_K_M) requires 0.6 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run InternVL2 1B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 6.0 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~797Simulate output on RTX 3080RTX 2080 Ti~646Simulate output on RTX 2080 TiRTX 3070 Ti~638Simulate output on RTX 3070 TiM3*8 GB~116Simulate output on iMac 24" M3 (8 GB)Orin Nano~107Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~77Simulate output on iPhone 1720 GB
fits · 18.0 GB free · up to 32K ctx
1 GPU
RX 7900 XT~774Simulate output on RX 7900 XT32–40 GB
fits · 30.0 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~1879Simulate output on RTX 5090M5 Max36 GB~519Simulate output on Mac Studio M5 Max (36 GB)M532 GB~173Simulate output on MacBook Air 13" M5 (32 GB)Orin~215Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~142Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
2| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.5 0.8BAlibaba | — | 0.9 GB | ~753 | same size |
| Gemma 3 1B ITGoogle | — | 2.1 GB | ~993 | same size |
Specifications
- Publisher
- OpenGVLab
- Parameters
- 938M
- Architecture
- InternVLChatModel
- Release Date
- 2024-07-08
- License
- MIT
Frequently Asked Questions
How much VRAM does InternVL2 1B need?
InternVL2 1B requires 0.6 GB of VRAM at Q4_K_M, or 2.1 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 938M × 4.8 bits ÷ 8 = 0.6 GB
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 InternVL2 1B?
For InternVL2 1B, Q4_K_M (0.6 GB) offers the best balance of quality and VRAM usage. Q5_K_S (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 InternVL2 1B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run InternVL2 1B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.6 GB of usable unified memory (RAM minus macOS overhead).
Can I run InternVL2 1B locally?
Yes — InternVL2 1B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is InternVL2 1B?
At Q4_K_M, InternVL2 1B can reach ~7742 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~1057 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 ÷ 0.62 × 0.65 = ~8387 tok/s
Estimated speed at Q4_K_M (0.6 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of InternVL2 1B?
At Q4_K_M, the download is about 0.56 GB. The full-precision BF16 version is 1.88 GB. The smallest option (IQ2_XXS) is 0.26 GB.
Which GPUs can run InternVL2 1B?
52 consumer GPUs can run InternVL2 1B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run InternVL2 1B?
163 devices with unified memory can run InternVL2 1B at Q4_K_M (0.6 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.