The Calculator - VRAMCalculator.com

What LLM Can My Mac Run? Check Your Apple Silicon in Seconds

October 25, 2026August 21, 2026 by The Calculator

VRAMCalculator.com ALPHA build 0.1 What LLM Can My Mac Run? Usable by the GPU 0 GB Chip generation M1 M2 M3 M4 M5 Your Mac Number of machines GPU memory limit 75% of RAM macOS reserves the rest for the system. The default is about 75%, and some machines report less. Run ggml_metal_init in llama.cpp … Read more

Ornith 1.5 9B on an RX 580: 16.3 tok/s, and It Wrote This

August 21, 2026August 20, 2026 by The Calculator

rx 580 llm, ornith 1.5 vram, local llm write article, intel arc llama.cpp, vulkan llama.cpp speed

Ornith 1.5 9B Runs at 16.3 tok/s on a Card From 2017 We put Ornith 1.5 9B on a Radeon RX 580, a nine year old 8 GB card, and it generated at 16.3 tokens per second at Q4_K_M with a 32,000 token context. On a 2024 Intel laptop it managed 8.6. Then we handed … Read more

DFlash 2 Gets Quoted at 90 tok/s. The Table Says Less.

September 23, 2026August 19, 2026 by The Calculator

dflash2, qwen 3.8 27b dflash, block diffusion drafting, speculative decoding llama.cpp, mtp vs dflash

Does DFlash 2 Actually Double Qwen 3.8 27B? Yes, against a plain baseline. One A100 test went from 28.9 to 59.1 tok/s, and one RTX 4090 test went from 60 tok/s on MTP to about 90 on DFlash 2. The word “again” matters. DFlash 2 is not a faster MTP. It is a second, independent … Read more

Qwen 3.8 27B Ties GPT-5.6: The Small Model Distillation Era

August 18, 2026 by The Calculator

How Does a 27B Model Tie GPT-5.6? Qwen 3.8 27B scores 52 on the Artificial Analysis Intelligence Index, the same score as GPT-5.6 Luna (max), and one point behind GLM-5.2 at 753B and DeepSeek V4 Pro at 1.6T. The numbers come from Artificial Analysis, measured on their nine-eval index, and reported on 17 August 2026 … Read more

Abliterated Models: Meaning, vs Uncensored, VRAM Needs

October 7, 2026August 16, 2026 by The Calculator

abliterated models, abliterated models huggingface, abliterated model meaning, uncensored vs abliterated, abliterated model vram, how to run abliterated models

What Is an Abliterated Model? An abliterated model is an open-weight LLM whose refusal behavior has been surgically removed by deleting a single direction from its weights, rather than by retraining. The word blends “ablation”, the neuroscience term for cutting out a function to see what it does, with “obliterated.” The result answers prompts the … Read more

Qwen3.8 27B Speed: Why the Same Model Runs 5 or 144 tok/s

October 4, 2026August 15, 2026 by The Calculator

qwen3.8 27b tokens per second, qwen 3.8 27b tok/s, qwen3.8 27b 3090, qwen3.8 27b 4090 speed, qwen3.8 27b mtp, qwen3.8 27b benchmark, how fast is qwen3.8 27b, qwen3.8 27b mac

Qwen3.8 27B Speed on Real Hardware, Measured by the People Who Ran It Qwen3.8 27B runs at roughly 5 to 144 tokens per second depending on the hardware, and the biggest single factor is not the card: it is whether multi-token prediction is switched on. An RTX 5090 goes from 66 to 144 with one … Read more

Nemotron 3.5 Lightning VRAM Requirements: The Quant Floor

August 15, 2026August 12, 2026 by The Calculator

nemotron 3.5 lightning vram, nemotron 3.5 lightning 30b, nemotron 3.5 lightning gguf, can i run nemotron 3.5 lightning, nemotron 30b a3b vram, nvidia nemotron 3.5 lightning 24gb

How Much VRAM Does Nemotron 3.5 Lightning Need? About 20.1 GB at the smallest published builds, and about 26.6 GB at Q4_K_M, including an 8K cache and llama.cpp overhead. Nemotron 3.5 Lightning is NVIDIA’s 30B-A3B Mamba hybrid, published on August 11, 2026 with roughly 3.6B parameters active per token. The number most people will want … Read more

Qwen3.8 VRAM: the 27B That Ties GPT-5.6 Fits 24GB

October 4, 2026August 12, 2026 by The Calculator

qwen3.8-27b vram, qwen3.8-max, qwen3.8 open weights, qwen3.8 benchmarks, run qwen3.8 locally, qwen3.8 27b requirements

Qwen3.8 VRAM Requirements: Both Models Now Measured Qwen3.8-27B weights went public at 15:00 UTC on 14 August 2026, two days after the 2.4T, and on Alibaba’s own comparison table it trades blows with Opus 4.6 Max. Both halves of this page are now measurement rather than projection, and the 27B figures below come from unsloth’s … Read more

Muse Glimmer VRAM Requirements: Three Files, Not One

September 23, 2026August 10, 2026 by The Calculator

muse glimmer vram, muse glimmer 30b, can i run muse glimmer, muse glimmer gguf, muse glimmer 24gb, meta muse glimmer local, muse glimmer mmproj, muse glimmer dflash

How Much VRAM Does Muse Glimmer Need? About 17.1 GB for the Q4 weights and a short context, and about 21.9 GB if you want vision, the speculative decoding drafter and the full 128K context at the same time. Muse Glimmer is a 29.6B dense multimodal model from Meta Superintelligence Lab, published as meta-models/Muse-Glimmer-30B on … Read more

QLoRA VRAM Requirements: Why Every Number Disagrees

August 7, 2026 by The Calculator

QLoRA VRAM Requirements: Why Every Number You Find Is Different How Much VRAM Does QLoRA Need? For a 7B model, published answers range from 5 GB to 20 GB, and all of them are defensible. Unsloth says 5 GB. Modal says 5 GB. Spheron says about 8 GB. RunPod says 8 to 10 GB. Introl … Read more