Nemotron Cascade 2 30B A3B JANG 2L CRACK — Hardware Requirements & GPU Compatibility
Chat
Nemotron Cascade 2 30B A3B JANG 2L CRACK is a 36.2B-parameter open language model from dealignai in the Nemotron family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 22.11 GB of VRAM — see which GPUs and Macs can run it below.
Runs well on a 24 GB GPU
Runs from 16 GB (Q2_K · 32K)Best on 24 GB (Q4_K_M · ~177 tok/s · 256K)Macs from 24 GB (Q3_K_M · ~58 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Nemotron Cascade 2 30B A3B JANG 2L CRACK
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve dealignai/Nemotron-Cascade-2-30B-A3B-JANG_2L-CRACK --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A6000 (48 GB) (25 GB headroom) · Q4_K_M
- Generation speed
- ~153 tok/s
- generation speed
- Cost per 1M output tokens
- $0.60
- per 1M output tokens
How Much VRAM Does Nemotron Cascade 2 30B A3B JANG 2L CRACK Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 15.8 GB | 15.9 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 256K context · ~206 tok/s | |
| 18.0 GB | 18.1 GB | 3-bit medium — 3-bit medium quantization | Fits: 256K context · ~195 tok/s | |
| 22.1 GB | 22.2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 256K context · ~177 tok/s | |
| 26.2 GB | 26.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Does not fit: needs 26.2 GB | |
| 30.3 GB | 30.4 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 30.3 GB | |
| 36.6 GB | 36.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 36.6 GB | |
| 72.8 GB | 72.9 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 72.9 GB |
Selected Q4_K_M
4-bit medium quantization22.1GBat 16K context
- Weights
- 21.7 GB
- KV cache
- 0.1 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 1.9 GB free · ~177 tok/s
≥50 tok/s needs ≥164 GB/s · RTX 5090 ~226, RTX 3090 Ti ~177
Which Hardware Can Run Nemotron Cascade 2 30B A3B JANG 2L CRACK?
Q4_K_M · 22.1 GBNemotron Cascade 2 30B A3B JANG 2L CRACK (Q4_K_M) requires 22.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 29+ GB is recommended. Using the full 262K context window can add up to 1.2 GB, bringing total usage to 23.3 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090. 100 devices with unified memory can run Nemotron Cascade 2 30B A3B JANG 2L CRACK, including NVIDIA DGX Spark, NVIDIA Jetson AGX Thor Developer Kit, Mac Studio M5 Max (36 GB).
20 GB
Q4_K_M · best fit Q3_K_M (18.1 GB)
1 GPU
RX 7900 XT~74Simulate output on RX 7900 XT32–40 GB
fits · 9.9 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~226Simulate output on RTX 5090M5 Max36 GB~62Simulate output on Mac Studio M5 Max (36 GB)M532 GB~37Simulate output on MacBook Air 13" M5 (32 GB)Orin~60Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~38Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Nemotron Cascade 2 30B A3B JANG 2L CRACK
Nemotron family · 42 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.6 35B A3BAlibaba | 106.8 | 22 GB | ~151 | same size |
| MiniCPM V 4.7 35B A3BOpenBMB | — | 22.1 GB | ~169 | same size |
| Ornith 1.5 35B A3Bornith-ai | — | 22 GB | ~151 | same size |
| Agents A1InternScience | — | 21.5 GB | ~171 | same size |
| Intern S2 PreviewInternLM | — | 73.2 GB | —needs > 24 GB | same size |
| EXAONE 4.5 33BLGAI-EXAONE | — | 23.9 GB | ~33 | same size |
Specifications
- Publisher
- dealignai
- Family
- Nemotron
- Parameters
- 36.2B
- Active parameters
- 3B per token
- Architecture
- NemotronHForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2026-03-22
- License
- Other
Frequently Asked Questions
How much VRAM does Nemotron Cascade 2 30B A3B JANG 2L CRACK need?
Nemotron Cascade 2 30B A3B JANG 2L CRACK requires 22.1 GB of VRAM at Q4_K_M, or 72.8 GB at BF16. Full 262K context adds up to 1.2 GB (23.3 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 36.2B × 4.8 bits ÷ 8 = 21.7 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 ≈ 1.6 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Nemotron Cascade 2 30B A3B JANG 2L CRACK?
Yes, at Q4_K_M (22.1 GB) or lower. Higher quantizations like Q5_K_M (26.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Nemotron Cascade 2 30B A3B JANG 2L CRACK?
For Nemotron Cascade 2 30B A3B JANG 2L CRACK, Q4_K_M (22.1 GB) offers the best balance of quality and VRAM usage. Q5_K_M (26.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 15.8 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Nemotron Cascade 2 30B A3B JANG 2L CRACK on a Mac?
Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Nemotron Cascade 2 30B A3B JANG 2L CRACK, the smallest at Q3_K_M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 22.1 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).
Can I run Nemotron Cascade 2 30B A3B JANG 2L CRACK locally?
Yes — Nemotron Cascade 2 30B A3B JANG 2L CRACK can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Nemotron Cascade 2 30B A3B JANG 2L CRACK?
At Q4_K_M, Nemotron Cascade 2 30B A3B JANG 2L CRACK can reach ~93 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~177 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 1.8 GB active ÷ (8000 × 0.65) = 0.352 ms, plus 52 layers × 0.055 ms = 2.860 ms, so 1000 ÷ 3.212 ms = ~311 tok/s
Estimated speed at Q4_K_M (22.1 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Nemotron Cascade 2 30B A3B JANG 2L CRACK?
At Q4_K_M, the download is about 21.74 GB. The full-precision BF16 version is 72.47 GB. The smallest option (Q2_K) is 15.40 GB.
Which GPUs can run Nemotron Cascade 2 30B A3B JANG 2L CRACK?
7 consumer GPUs can run Nemotron Cascade 2 30B A3B JANG 2L CRACK at Q4_K_M. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.
Which devices can run Nemotron Cascade 2 30B A3B JANG 2L CRACK?
100 devices with unified memory can run Nemotron Cascade 2 30B A3B JANG 2L CRACK at Q4_K_M (22.1 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.