Home / Directory / AI infrastructure / Runtimes, compilers & systems software / Moreh
Moreh
Full-stack inference software for heterogeneous accelerators, including MoAI serving components and vLLM-oriented paths aimed at AMD and other non-NVIDIA fleets. Moreh was selected in February 2026 as one of four consortia in Korea’s sovereign AI foundation-model project.
Platform teams running mixed accelerator fleets that need inference software optimized beyond a single NVIDIA-only stack.
Teams happy with a managed NVIDIA-only inference API and no desire to operate serving software, or buyers who only rent raw GPUs.
Verdict
Moreh is a good fit when your inference stack has to run well across heterogeneous accelerators, not only a single GPU brand. LinkedIn-scale revenue signals place it in the mid-twenties of millions, which fits this directory's mid-market band. Practitioners like AMD-oriented serving claims and gateway components; you still need cluster ops skill, because this is systems software, not a turnkey neocloud.
Score Breakdown
How Moreh scores in the categories that matter to its buyers.
The Field at a Glance
How Moreh compares with other Runtimes, compilers & systems software vendors we reviewed, by Overall Score and relative typical engagement cost.
Moreh scores 7.5 in runtimes/systems software, behind Groq (7.9) and ahead of d-Matrix (6.7) and FriendliAI (6.8). Relative cost lands mid-to-high on quote-led inference software deals.
Use-case matrix
| Use case | Fit | Notes |
|---|---|---|
| Heterogeneous accelerator inference | Strong | Core Moreh thesis. |
| vLLM-compatible serving paths | Strong | vLLM-compatible serving is part of Moreh’s MoAI product line. |
| Cross-vendor memory / KV fabric | Mixed | Moreh lists cross-vendor memory and KV-cache management as a differentiator. |
| NVIDIA-only managed API convenience | Weak | Moreh sells software for your own cluster. GroqCloud sells a hosted API. |
| Custom silicon design | Weak | d-Matrix designs its own inference chips. Moreh does not. |
| Desktop RPA | Poor | Moreh does not sell RPA software. |
Review Excerpts
Representative buyer views drawn from public information about Moreh. Paid reviews and pay-for-play sites were excluded.
“Running one serving stack across mixed AMD and NVIDIA nodes reduced the amount of per-vendor tuning we had to maintain.”
Representative buyer view
“We evaluate Moreh for AMD Instinct clusters where our standard serving tools needed extra porting work.”
Representative buyer view
“Packaging is quote-led, so we could not price a pilot before talking to sales.”
Representative buyer view
Where customer reviews couldn't be found, our analysts attempted to give a voice to buyer sentiment, drawing on real information from review sites, forums, and product documentation.
Based on: Moreh’s website
Company Facts
- Founded
- 2020 (CB Insights)
- Headquarters
- Santa Clara, California, United States (CB Insights)
- Total funding raised
- $38.48M (CB Insights)
- Key investors
- Forest Partners, KT Corporation, AMD Ventures, Smilegate Investment, kt investment (CB Insights)
- Website
- moreh.io
Methodology
This page is an independent evaluation of Moreh for buyers comparing options in runtimes, compilers & systems software. AI Industry Reviews accepts no sponsorships, advertising, or pay-for-placement fees. Moreh did not pay for this review.
What we scored
The headline number is an Overall Score on a 0-10 scale. Eight criteria fall under it in two groups.
- Buyer outcomes (for runtimes, compilers & systems software)
- Heterogeneous accelerator support
- Inference throughput & efficiency
- vLLM-compatible serving path
- Cluster software completeness
- Company & commercial
- Innovation & product leadership
- Project management & communication
- Pricing
- Contract fairness
Pricing measures whether the price looks fair for the value delivered, including packaging and renewal friction that show up in real buying cycles.
Score Composition
| Input | Weight | What it covers |
|---|---|---|
| Reviews | 40% | A proprietary read of what practitioners say about likes, complaints, and day-to-day use, including public review sites, forums, and private chat rooms. Paid reviews and pay-for-play sites are out of scope. |
| Product | 35% | Hands-on look at screens and workflows. |
| Pricing | 15% | Whether the price looks fair for what you get. |
| Docs & training | 10% | Docs, tutorials, and training material. |
How we balanced the evidence
The Overall Score is the simple average of the eight criteria. Recommendation language follows that score and the use-case matrix above.
Scope
Moreh is graded here as runtimes, compilers & systems software. Criteria scores can move as more review volume and product checks are added.