OpenJarvis: Personal AI, On Personal Devices
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918507713658880 |
|---|---|
| author | Saad-Falcon, Jon Narayan, Avanika Manihani, Robby Bhathal, Tanvir Shandilya, Herumb Akengin, Hakki Orhun Bo, Gabriel Park, Andrew Hart, Matthew Costello, Caia Li, Chuan Ré, Christopher Mirhoseini, Azalia |
| author_facet | Saad-Falcon, Jon Narayan, Avanika Manihani, Robby Bhathal, Tanvir Shandilya, Herumb Akengin, Hakki Orhun Bo, Gabriel Park, Andrew Hart, Matthew Costello, Caia Li, Chuan Ré, Christopher Mirhoseini, Azalia |
| contents | Personal AI stacks, like OpenClaw and Hermes Agent, are becoming central to daily work, yet they route nearly every query (often over sensitive local data) to cloud-hosted frontier models. Replacing frontier models with local models inside existing stacks does not work: swapping Claude Opus 4.6 for Qwen3.5-9B drops accuracy by 25-39 pp across personal AI tasks like PinchBench and GAIA. Existing stacks bundle agentic prompts, tool descriptions, memory configuration, and runtime settings around a specific cloud model. Only the prompts can be tuned, and state-of-the-art prompt optimizers close just 5 pp of the local-cloud gap on their own. This motivates a decomposed personal AI stack: one that exposes individual primitives which can be optimized individually or jointly to close the local-cloud gap. We present OpenJarvis, an architecture that represents a personal AI system as a typed spec over five primitives: Intelligence, Engine, Agents, Tools & Memory, and Learning. Each primitive is an independently editable field, making the stack end-to-end optimizable and measurable against accuracy, cost, and latency. Towards closing the local-cloud gap without surrendering local-model properties, OpenJarvis introduces LLM-guided spec search, a local-cloud collaboration in which frontier cloud models propose edits across the spec at search time, only non-regressing edits are accepted, and the resulting spec runs entirely on-device at inference time. With LLM-guided spec search, on-device specs match or exceed cloud accuracy on 4 of 8 benchmarks and land within 3.2 pp of the best cloud baseline on average. They also reduce marginal API cost by ~800x and end-to-end latency by 4x. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_17172 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OpenJarvis: Personal AI, On Personal Devices Saad-Falcon, Jon Narayan, Avanika Manihani, Robby Bhathal, Tanvir Shandilya, Herumb Akengin, Hakki Orhun Bo, Gabriel Park, Andrew Hart, Matthew Costello, Caia Li, Chuan Ré, Christopher Mirhoseini, Azalia Machine Learning Artificial Intelligence Computation and Language Personal AI stacks, like OpenClaw and Hermes Agent, are becoming central to daily work, yet they route nearly every query (often over sensitive local data) to cloud-hosted frontier models. Replacing frontier models with local models inside existing stacks does not work: swapping Claude Opus 4.6 for Qwen3.5-9B drops accuracy by 25-39 pp across personal AI tasks like PinchBench and GAIA. Existing stacks bundle agentic prompts, tool descriptions, memory configuration, and runtime settings around a specific cloud model. Only the prompts can be tuned, and state-of-the-art prompt optimizers close just 5 pp of the local-cloud gap on their own. This motivates a decomposed personal AI stack: one that exposes individual primitives which can be optimized individually or jointly to close the local-cloud gap. We present OpenJarvis, an architecture that represents a personal AI system as a typed spec over five primitives: Intelligence, Engine, Agents, Tools & Memory, and Learning. Each primitive is an independently editable field, making the stack end-to-end optimizable and measurable against accuracy, cost, and latency. Towards closing the local-cloud gap without surrendering local-model properties, OpenJarvis introduces LLM-guided spec search, a local-cloud collaboration in which frontier cloud models propose edits across the spec at search time, only non-regressing edits are accepted, and the resulting spec runs entirely on-device at inference time. With LLM-guided spec search, on-device specs match or exceed cloud accuracy on 4 of 8 benchmarks and land within 3.2 pp of the best cloud baseline on average. They also reduce marginal API cost by ~800x and end-to-end latency by 4x. |
| title | OpenJarvis: Personal AI, On Personal Devices |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2605.17172 |