exo
Hacker ApprovedShares rank 6 of 27 inference engines by recommenders
People recommend exo for pooling everyday devices, such as Mac Studios and DGX Sparks, into one cluster for running large local AI models. Its slow single-request speed is the main caveat.
People recommend exo for pooling everyday devices, such as Mac Studios and DGX Sparks, into one cluster for running large local AI models. Its slow single-request speed is the main caveat.
18 people gave an opinion
- recommend it
- 15
- criticize it
- 3
- approval share
- 83%
Recommended in 9 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 15 people recommend itNeeds 10 or more, met.
- 5 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 9 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Pools devices into one AI cluster
“You can do local AI inference and get Claude Opus-level performance (Kimi K2.5) over a cluster of Mac Studios with Exo.Labs”
Comment from 20 Feb 2026 in “I tried building my startup entirely on European infrastructure” “That's why Exo coupled DGX Spark (5090 performance for FP4) with MacStudio and sped it up 4x.”
Comment from 19 Dec 2025 in “1.5 TB of VRAM on Mac Studio – RDMA over Thunderbolt 5” “EXO lets you run your own AI cluster at home with everyday devices.”
Comment from 18 Dec 2025 in “EXO v1 Release” - Comment from 1 Dec 2025 in “DeepSeek-v3.2: Pushing the frontier of open large language models [pdf]”
- Comment from 15 Jan 2026 in “Apple Is Fighting for TSMC Capacity as Nvidia Takes Center Stage”
- Comment from 16 Feb 2026 in “Qwen3.5: Towards Native Multimodal Agents”
Strong fit with Apple Silicon
“Exo, and the teams at Apple that built the foundations Exo uses, actually use modern distributed compute.”
Comment from 13 Oct 2026 in “A Dynamic and Distributed Virtual GPU Cloud Built on Idle Apple Silicon Devices” “the insanse community work on Apple Silicon across whisper.cpp/mflux/llama.cpp/MLX/Exo”
Comment from 11 Oct 2026 in “LineageOS 24” “between Apple Silicon and Exo there are some amazingly solutions out there”
Comment from 18 Apr 2026 in “Opus 4.7 to 4.6 Inflation is ~45%”
Other inference engines people recommend
The four most recommended other inference engines, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| vLLM | 112 | 56 | 67% | 12 | — | Approved |
| Strata | 31 | 15 | 67% | 5 | — | Approved |
| DwarfStar 4 | 26 | 2 | 93% | 6 | — | Approved |
| SGLang | 19 | 13 | 59% | 8 | — | — |
| exoThis page | 15 | 3 | 83% | 9 | — | Approved |
What people said
16 of the 21 comments that mention it, newest first. Each links to the original comment.
“the penalties were brutal”
“Exo, and the teams at Apple that built the foundations Exo uses, actually use modern distributed compute.”
“the insanse community work on Apple Silicon across whisper.cpp/mflux/llama.cpp/MLX/Exo”
“I ended up switching that particular part to Exo”
“has the potential to complete tasks with fewer turns and cost”
“architecturally it separates things in a way that makes it a lot harder for the agent to break itself when modifying itself”
“with exo I can run decent quantization with 1 millon tokens for memory and derailment tests.”
“between Apple Silicon and Exo there are some amazingly solutions out there”
“You can do local AI inference and get Claude Opus-level performance (Kimi K2.5) over a cluster of Mac Studios with Exo.Labs”
“stream it to M3 using Exo”
“connect them together using Exo”
“That's why Exo coupled DGX Spark (5090 performance for FP4) with MacStudio and sped it up 4x.”
“Exo just outputs a time in ms, no other data for that”
“EXO lets you run your own AI cluster at home with everyday devices.”
“its slow but scales linearly with concurrent requests”
“hook a DGX Spark via Exo Labs stack”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.