Intern Decision 2B — Hardware Requirements & GPU Compatibility
VisionFunctionsIntern-Decision-2B is InternLM's 2.2-billion-parameter multimodal structured decision model, fine-tuned from Qwen3.5-2B. Instead of generating free-form text, it accepts a shared state, a schema of named questions with optional images, and returns an answer distribution for each question in one forward pass by scoring candidate option symbols. The card shows average scores of 84.68 on decision benchmarks and roughly 33 ms per query using a single RTX 4090 GPU. At this scale, it runs easily on modest consumer hardware. The model handles context windows up to 262,144 tokens. Released under the Apache 2.0 license, it allows unrestricted commercial and research use and appeared in September 2026. Positioned between the 0.8B and 4B models within the Intern-Decision family, using it requires the custom inference code provided with the repository instead of a standard chat interface.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 256K)Best on 24 GB (Q4_K_M · ~379 tok/s · 256K)Macs from 8 GB (Q4_K_M · ~41 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Intern Decision 2B
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/bombdefuser-124/Intern-Decision-2B-GGUFQ4_K_M fits fully on the RTX 4090 · 22.2 GB free · 16K context
Download: Intern-Decision-2B-Q4_K_M.gguf1.33 GBbombdefuser-124
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (10 GB headroom) · Q4_K_M
- Generation speed
- ~135 tok/s
- generation speed
- Cost per 1M output tokens
- $0.13
- per 1M output tokens
How Much VRAM Does Intern Decision 2B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.3 GB | 1.6 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~489 tok/s | |
| 1.5 GB | 1.8 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~443 tok/s | |
| 1.7 GB | 2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~379 tok/s | |
| 2.0 GB | 2.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 256K context · ~331 tok/s | |
| 2.2 GB | 2.5 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 256K context · ~294 tok/s | |
| 2.6 GB | 2.9 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 256K context · ~251 tok/s | |
| 4.8 GB | 5.1 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 256K context · ~136 tok/s |
Selected Q4_K_M
4-bit medium quantization1.8GBat 16K context
- Weights
- 1.3 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Which Hardware Can Run Intern Decision 2B?
Q4_K_M · 1.7 GBIntern Decision 2B (Q4_K_M) requires 1.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 262K context window can add up to 3.1 GB, bringing total usage to 4.8 GB. 114 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 185 devices with unified memory can run Intern Decision 2B, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB).
≤ 11 GB
≤ 11 GB tier →fits · 2.2 GB free · up to 128K ctx
68 GPUs · 1 Mac class · 3 device families
RTX 3080~286Simulate output on RTX 3080RTX 2080 Ti~231Simulate output on RTX 2080 TiRTX 3070 Ti~229Simulate output on RTX 3070 TiM3*8 GB~41Simulate output on iMac 24" M3 (8 GB)Orin Nano~38Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~28Simulate output on iPhone 1720 GB
fits · 18.2 GB free · up to 256K ctx
1 GPU
RX 7900 XT~278Simulate output on RX 7900 XT32–40 GB
fits · 30.2 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 9 device families
RTX 5090~673Simulate output on RTX 5090M5 Max*36 GB~186Simulate output on Mac Studio M5 Max (36 GB)M632 GB~69Simulate output on iMac 24" M6 (32 GB)RTX Spark N1X~113Simulate output on NVIDIA RTX Spark N1X Laptop (32 GB)Orin~77Simulate 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 |
|---|---|---|---|---|
| LFM3 Omni 2BLiquid AI | — | 2.4 GB | ~385 | same size |
| Qwen3.5 2BAlibaba | — | 1.9 GB | ~370 | same size |
| Qwen3 VL 2B InstructAlibaba | — | 3.5 GB | ~362 | same size |
| Qwen2 VL 2B InstructAlibaba | — | 2.1 GB | ~390 | same size |
| Kimi K3 DSparkRadixArk | — | 2.2 GB | ~381 | same size |
| Gemma 4 31B IT Speculator.eagle3RedHatAI | — | 5 GB | ~205 | same size |
Specifications
- Publisher
- InternLM
- Parameters
- 2.2B
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-09-26
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Intern Decision 2B need?
Intern Decision 2B requires 1.7 GB of VRAM at Q4_K_M, or 4.8 GB at BF16. Full 262K context adds up to 3.1 GB (4.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 2.2B × 4.8 bits ÷ 8 = 1.3 GB
KV Cache + Overhead ≈ 0.4 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.
KV Cache + Overhead ≈ 3.6 GB (at full 262K context)
VRAM usage by quantization
What's the best quantization for Intern Decision 2B?
For Intern Decision 2B, Q4_K_M (1.7 GB) offers the best balance of quality and VRAM usage. Q5_K_M (2.0 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.3 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Intern Decision 2B on a Mac?
Yes — Mac Mini M2 (8 GB) and 158 other Macs can run Intern Decision 2B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 1.7 GB of usable unified memory (RAM minus macOS overhead).
Can I run Intern Decision 2B locally?
Yes — Intern Decision 2B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Intern Decision 2B?
At Q4_K_M, Intern Decision 2B can reach ~2775 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~379 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 ÷ 1.73 × 0.65 = ~3006 tok/s
Estimated speed at Q4_K_M (1.7 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Intern Decision 2B?
At Q4_K_M, the download is about 1.33 GB. The full-precision BF16 version is 4.43 GB. The smallest option (Q2_K) is 0.94 GB.
Which GPUs can run Intern Decision 2B?
114 consumer GPUs can run Intern Decision 2B 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 Intern Decision 2B?
185 devices with unified memory can run Intern Decision 2B at Q4_K_M (1.7 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.