Range – open a 1 TB AI model in 3 seconds without downloading it

Software engineer Andrey Grehov released Range, an open-source tool that opens a shell in a container image, Hugging Face repository, or S3/HTTP environment by reading only the bytes a program touches rather than downloading the whole source. In benchmarks run on EC2 m6i.large in us-east-1 on 28 September 2026, Range answered a llama.cpp chat prompt from the 6.38 GB unsloth/gemma-3-270m-it-GGUF repository in 6.7 seconds with the image indexed versus 18.3 seconds for docker pull plus hf download, and read a single tensor from the 1.03 TB, 61-shard moonshotai/Kimi-K2-Instruct model in 3.4 seconds while moving 9.5 MB of the model. Range installs as a macOS or Linux release archive (x86-64 or arm64), requires Lima for its Linux VM on macOS, and needs root plus the nbd, erofs and overlay kernel modules on Linux.

README.md Range opens a shell in a container image, a Hugging Face repository, or an environment in S3 or on any HTTP server, without downloading it first. Only the bytes your program reads cross the network. bash $ range shell python:3.12 $ range shell python:3.12 --mount hf://moonshotai/Kimi-K2-Instruct:/model $ range shell s3://<your-bucket /dev.range No Docker, no daemon and no pull. Linux runs it natively. On macOS, Range runs Linux in a small VM that it manages itself. Measured on EC2 in us‑east‑1, with the image indexed once. See bench.log. demo.txt bash $ range run ghcr.io/ggml-org/llama.cpp:light-b11206 \ --mount hf://unsloth/gemma-3-270m-it-GGUF:/model -- \ llama-cli -m /model/gemma-3-270m-it-Q4 K M.gguf -st \ -p "Why is the sky blue? Answer in one sentence." The sky is blue because of a phenomenon called Rayleigh scattering, where blue light is scattered more than other colors. Range opens the llama.cpp image from its registry and mounts the model repository at /model . The repository holds 6.38 GB in 24 files. Range reads one of them. With the image indexed, the answer took 6.7 s from an empty cache. docker pull plus hf download took 18.3 s. The very first run, which also indexes the image, took 15.5 s. bash $ range run python:3.12 --mount hf://moonshotai/Kimi-K2-Instruct:/model -- \ du -sh --apparent-size /model 959G /model Kimi K2 is 1.03 TB in 61 shards. A Python script inside read its config, the header of one shard and one tensor. That took 3.4 s with the image indexed, and moved 9.5 MB of the model. The other 60 shards never left Hugging Face. Range reads a file when a program opens it. A program that reads a whole model still downloads the whole model, once. bench.log | Empty cache to output | docker pull | Range, first run | Range, indexed | Range, again | |---|---|---|---|---| | Chat demo | 18.3 s 579 MB | 15.5 s 590 MB | 6.7 s 317 MB | 4.2 s 0 MB | | python:3.12 | 15.8 s 435 MB | 16.5 s 415 MB | 2.8 s 48 MB | 1.1 s 0 MB | | rust:1.82 | 19.6 s 569 MB | 22.4 s 546 MB | 8.0 s 125 MB | 1.6 s 0 MB | | eclipse-temurin:21 | 7.9 s 232 MB | 9.2 s 225 MB | 2.9 s 49 MB | 1.0 s 0 MB | | Kimi K2, 1 TB | not tried 1.03 TB | 17.6 s 433 MB | 3.4 s 56 MB | 1.2 s 0 MB | The commands: import json and sqlite3, cargo --version, java -version, and a read of one Kimi K2 tensor. A first run reads each layer once to index it. The python:3.12 index is 4.1 MB. Medians of three, m6i.large, us‑east‑1, 28 September 2026. Every run starts empty, except "again". The bars replay at 3x speed. problem.txt A machine that needs a large environment downloads all of it, every time, to use a small part. Range reads only the bytes each machine touches, and the next run fetches them before it asks. design.txt ReadAt offset, length - bytes . Range turns a source into a disk, and turns each read of that disk into a ranged request to the source. install.txt A release archive for macOS or Linux, x86-64 or arm64. On macOS, Range also needs Lima for its Linux VM. Then open a shell in any image: bash $ curl -fsSL https://github.com/andreygrehov/range/releases/latest/download/range $ uname -s $ uname -m .tar.gz | tar -xz $ brew install lima macOS only $ ./range shell python:3.12 For your own environments, build once, and every first run reads lazily: bash $ range build --from-oci python:3.12 -o py.range $ range publish py.range s3://<your-bucket /py.range $ range shell s3://<your-bucket /py.range A published artifact needs no indexing. The go1.23 demo artifact was ready in 0.41 s on its first run, and moved 6 MB of 1.03 GB. Or build from source, with Go 1.25 or newer: bash $ git clone https://github.com/andreygrehov/range && cd range && make install Linux needs root and the nbd, erofs and overlay kernel modules. range doctor checks them. Windows works through WSL2, untested. about me.txt I am a software engineer at AWS. Range is my personal project.