InternVL3 5 GPT OSS 20B A4B Preview HF — Hardware Requirements & GPU Compatibility
Vision
InternVL3.5-GPT-OSS-20B-A4B-Preview-HF is OpenGVLab's multimodal mixture-of-experts model with about 21.2 billion total parameters and 4 billion active, using OpenAI's gpt-oss-20b as the language model and the InternViT-300M vision encoder. It is a preview in the InternVL3.5 family, in the Transformers-native HF format. The card says that lmdeploy does not yet support GPT-OSS and recommends vLLM for deployment. With only 4 billion parameters active, it is relatively quick, and once quantized it fits on a single high-end consumer GPU or a machine with 16 GB or more of memory. It is released under the Apache 2.0 license, permitting commercial and research use, though the gpt-oss base model is a separate release. Published in August 2025, it is the mixture-of-experts variant of the family that otherwise uses Qwen3-based dense language models.
561.1K downloads
9 likesSpecifications
- Publisher
- OpenGVLab
- Family
- GPT-OSS
- Parameters
- 21.2B
- Architecture
- InternVLForConditionalGeneration
- Context Length
- 131,072 tokens
- Vocabulary Size
- 200,028
- Release Date
- 2025-08-29
- License
- Apache 2.0
Get Started
How Much VRAM Does InternVL3 5 GPT OSS 20B A4B Preview HF Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 9.4 GB | 13.8 GB | 9.02 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 10.7 GB | 15.2 GB | 10.35 GB | 3-bit medium quantization |
| Q4_K_Mest. | 4.80 | 13.1 GB | 17.6 GB | 12.74 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 15.5 GB | 20.0 GB | 15.13 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 17.9 GB | 22.4 GB | 17.52 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 21.6 GB | 26.1 GB | 21.23 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 42.8 GB | 47.3 GB | 42.47 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run InternVL3 5 GPT OSS 20B A4B Preview HF?
Q4_K_M · 13.1 GBInternVL3 5 GPT OSS 20B A4B Preview HF (Q4_K_M) requires 13.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 18+ GB is recommended. Using the full 131K context window can add up to 4.5 GB, bringing total usage to 17.6 GB. 26 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 5080.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run InternVL3 5 GPT OSS 20B A4B Preview HF?
Q4_K_M · 13.1 GB47 devices with unified memory can run InternVL3 5 GPT OSS 20B A4B Preview HF, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 (16 GB).
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does InternVL3 5 GPT OSS 20B A4B Preview HF need?
InternVL3 5 GPT OSS 20B A4B Preview HF requires 13.1 GB of VRAM at Q4_K_M, or 42.8 GB at BF16. Full 131K context adds up to 4.5 GB (17.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 21.2B × 4.8 bits ÷ 8 = 12.7 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 4.9 GB (at full 131K context)
VRAM usage by quantization
Q4_K_M13.1 GBQ4_K_M + full context17.6 GB- Can NVIDIA GeForce RTX 4090 run InternVL3 5 GPT OSS 20B A4B Preview HF?
Yes, at Q8_0 (21.6 GB) or lower. Higher quantizations like BF16 (42.8 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for InternVL3 5 GPT OSS 20B A4B Preview HF?
For InternVL3 5 GPT OSS 20B A4B Preview HF, Q4_K_M (13.1 GB) offers the best balance of quality and VRAM usage. Q5_K_M (15.5 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 9.4 GB.
VRAM requirement by quantization
Q2_K9.4 GBQ4_K_M ★13.1 GBQ5_K_M15.5 GBQ6_K17.9 GBQ8_021.6 GBBF1642.8 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run InternVL3 5 GPT OSS 20B A4B Preview HF on a Mac?
InternVL3 5 GPT OSS 20B A4B Preview HF requires at least 9.4 GB at Q2_K, 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 GPT OSS 20B A4B Preview HF locally?
Yes — InternVL3 5 GPT OSS 20B A4B Preview HF can run locally on consumer hardware. At Q4_K_M quantization it needs 13.1 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is InternVL3 5 GPT OSS 20B A4B Preview HF?
At Q4_K_M, InternVL3 5 GPT OSS 20B A4B Preview HF can reach ~186 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~180 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 ÷ 13.1 × 0.65 = ~539 tok/s
Estimated speed at Q4_K_M (13.1 GB)
~539 tok/s~180 tok/s~539 tok/s~456 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 GPT OSS 20B A4B Preview HF?
At Q4_K_M, the download is about 12.74 GB. The full-precision BF16 version is 42.47 GB. The smallest option (Q2_K) is 9.02 GB.
- Which GPUs can run InternVL3 5 GPT OSS 20B A4B Preview HF?
26 consumer GPUs can run InternVL3 5 GPT OSS 20B A4B Preview HF at Q4_K_M (13.1 GB). Top options include AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090, AMD Radeon RX 6800. 8 GPUs have plenty of headroom for comfortable inference.
- Which devices can run InternVL3 5 GPT OSS 20B A4B Preview HF?
49 devices with unified memory can run InternVL3 5 GPT OSS 20B A4B Preview HF at Q4_K_M (13.1 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (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.