VeriLoop E2 — Hardware Requirements & GPU Compatibility
ChatVisionFunctions
VeriLoop E2 is a 27.9B-parameter open language model from tsinghua-sigs-robot-lab. It supports a context window of up to 262,144 tokens. At BF16 it needs about 56.49 GB of VRAM — see which GPUs and Macs can run it below.
2.6K downloads
29 likesSpecifications
- Publisher
- tsinghua-sigs-robot-lab
- Parameters
- 27.9B
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-09-22
- License
- Apache 2.0
Get Started
HuggingFace
Run in cloud
Fits on A100 80GB (23 GB headroom) · BF16
- Generation speed
- ~24 tok/s
- generation speed
- Cost per 1M output tokens
- $12.80
- per 1M output tokens
How Much VRAM Does VeriLoop E2 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 56.5 GB | 70.4 GB | 55.75 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 VeriLoop E2?
BF16 · 56.5 GBVeriLoop E2 (BF16) requires 56.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 74+ GB is recommended. Using the full 262K context window can add up to 13.9 GB, bringing total usage to 70.4 GB. No consumer GPU has enough memory.
Which Devices Can Run VeriLoop E2?
BF16 · 56.5 GB22 devices with unified memory can run VeriLoop E2, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio (M3 Ultra, 96GB).
Runs great
— Plenty of headroom
Frequently Asked Questions
- How much VRAM does VeriLoop E2 need?
VeriLoop E2 requires 56.5 GB of VRAM at BF16. Full 262K context adds up to 13.9 GB (70.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 27.9B × 16 bits ÷ 8 = 55.7 GB
KV Cache + Overhead ≈ 0.8 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 ≈ 14.7 GB (at full 262K context)
VRAM usage by quantization
BF1656.5 GBBF16 + full context70.4 GB- Can NVIDIA GeForce RTX 5090 run VeriLoop E2?
No — VeriLoop E2 requires at least 56.5 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run VeriLoop E2 on a Mac?
Yes — Mac Studio M4 Max (64 GB) and 8 other Macs can run VeriLoop E2. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 56.5 GB of usable unified memory (RAM minus macOS overhead).
- Can I run VeriLoop E2 locally?
Yes — VeriLoop E2 can run locally on consumer hardware. At BF16 quantization it needs 56.5 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is VeriLoop E2?
At BF16, VeriLoop E2 can reach ~85 tok/s on AMD Instinct MI350X. 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 ÷ 56.5 × 0.65 = ~92 tok/s
Estimated speed at BF16 (56.5 GB)
~92 tok/s~92 tok/s~85 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of VeriLoop E2?
At BF16, the download is about 55.75 GB.
- Which GPUs can run VeriLoop E2?
No single consumer GPU has enough VRAM to run VeriLoop E2 at BF16 (56.5 GB). Multi-GPU or professional hardware is required.
- Which devices can run VeriLoop E2?
22 devices with unified memory can run VeriLoop E2 at BF16 (56.5 GB), including ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), 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.