jitllm
Runs LLM inference by compiling kernels for your CPUs and GPUs and paging model blocks from disk as needed.
not read yet
Use
- For
- people running LLMs on their own CPUs and GPUs, locally or on a server
- Price
- Price not read
- Activity
- Actively maintainedrelease 2 days ago
- Company
- Domain since 2026domain registered
- Runs as
- APISelf-hosted
- convert GGUF or Hugging Face weights into jlm files
- serve local models behind OpenAI-compatible and Anthropic-compatible APIs
- Domain since2026domain registered
Figure and bar: visits/mo, estimate · year: online since
- #1
Inferencer
Runs and inspects large language models locally with token-level control and distributed inference.
77ActiveFrom $9.99/mo - #2
mlx-serve
Runs open-source language models and generative tools locally on Mac, with chat, image generation, voice cloning, and agent capabilities.
2Active— - #3
Corbenic AI
Detects and skips redundant LLM computations to reduce inference costs by up to 98 percent.
149 linking domainsActiveFrom $5.12/mo - #4
Your Own AI
Runs large language models locally on your machine with optional cloud routing for frontier models.
60 linking domainsActiveFrom $200/yr - #5
Tensorant
Fine-tunes and serves open-weight models on your cloud GPUs with dataset management and evaluation.
Little sign of useActiveFrom $99/mo - #6
NVIDIAMCP · sign-in
Designs GPUs and AI computing platforms for data centers, gaming, and embedded systems.
12.6MActiveFrom ≈ $400 once1993
What is jitllm?
Runs LLM inference by compiling kernels for your CPUs and GPUs and paging model blocks from disk as needed. It is listed in the Atlas under LLM operations.
Who is jitllm for?
people running LLMs on their own CPUs and GPUs, locally or on a server
Do people use jitllm?
Emerging.
Is jitllm still maintained?
Actively maintained: release 2 days ago.
What are alternatives to jitllm?
In LLM operations, by use: Inferencer, mlx-serve, Corbenic AI, Your Own AI, Tensorant, NVIDIA.
Not quite it? Describe the job
betaFive tools from the Atlas that fit it, each judged against what you wrote.
How we read a tool
No votes, no reviews, no vendor claims. Every week a crawler reads each tool's own site and a few public registries, and the words on the card are bands over what it read.
- Use: visits to the site (estimated), installs from npm and PyPI, presence in Chrome's usage report.
- Activity: the newest release on GitHub, npm or PyPI; the newest dated page on the site; open roles on a public jobs board.
- Price: the vendor's own pricing page, read with the date. A figure is printed only when it is on that page.
- The line and the use cases are written by us from the homepage, one tool at a time, and refused when they repeat the vendor's marketing.
How AI assistants read each site — the reading vendors ask us about — is on each tool's own visibility page.