Home / Directory / AI infrastructure / Runtimes, compilers & systems software / d-Matrix
d-Matrix
In-memory compute inference accelerators plus the Aviator compiler and runtime stack for efficient large-model serving in the datacenter. d-Matrix raised a $275 million Series C at a $2 billion valuation in November 2025 and moved Corsair into full production in June 2026.
Inference platform teams evaluating in-memory accelerators with an MLIR-based compiler for lower energy per token at scale.
Developers who only want a hosted token API, or training-first GPU cloud buyers.
Verdict
d-Matrix fits when you are redesigning the inference rack around Corsair-class in-memory compute and the Aviator software stack, not when you merely need an API key. Full production announcements and compiler talks show a real systems software story beside the silicon. Commercial motion is quote-led hardware plus software, so bake off energy and throughput against GPUs before you commit. Skip it for self-serve token APIs or edge AIPUs.
Score Breakdown
How d-Matrix scores in the categories that matter to its buyers.
The Field at a Glance
Where d-Matrix ranks among Runtimes, compilers & systems software vendors we reviewed, by Overall Score and relative typical engagement cost.
d-Matrix scores 6.7, behind Groq (7.9) and FriendliAI (6.8). Its relative engagement cost is higher because the purchase is an accelerator and compiler program and not a token API.
Use-case matrix
| Use case | Fit | Notes |
|---|---|---|
| Datacenter inference acceleration | Strong | Corsair and Aviator are the core job. |
| Compiler / graph lowering for IMC | Strong | Aviator uses an MLIR-based compiler. |
| Hosted token API only | Poor | This is a hardware and software stack purchase. |
| Edge on-device AIPU | Poor | d-Matrix targets the datacenter. |
| GPU rental neocloud | Mixed | Corsair is often paired with GPUs, and d-Matrix does not rent them. |
| Training-first clusters | Poor | The design is built for inference efficiency. |
Review Excerpts
Excerpts from company materials and public coverage of d-Matrix, plus representative buyer views drawn from public information. Paid reviews and pay-for-play sites were excluded.
“Corsair entered full production to meet customer demand, with Aviator integrated into rack systems for GPU-plus-accelerator decode inference.”
“Aviator includes model adaptation, compression, an MLIR-based compiler, and a distributed inference engine with Kubernetes-oriented tooling.”
“Keeping weights next to the compute cut decode latency in our tests, which is why we shortlisted d-Matrix.”
Representative buyer view
“Adopting a new inference accelerator means compiler bring-up and ops training. Teams that expect drop-in CUDA containers will underestimate the integration cost.”
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: d-Matrix’s website and d-Matrix’s Series C announcement
Company Facts
- Founded
- 2019 (CB Insights)
- Headquarters
- Santa Clara, California, United States (CB Insights)
- Total funding raised
- $429M (CB Insights)
- Key investors
- NVIDIA, M12, Nautilus Venture Partners, Mirae Asset Financial Group, Temasek (CB Insights)
- Website
- d-matrix.ai
Key Leadership
Sid Sheth
Co-founder, President and CEO
Methodology
This page is an independent evaluation of d-Matrix for buyers comparing options in runtimes, compilers & systems software. AI Industry Reviews accepts no sponsorships, advertising, or pay-for-placement fees. d-Matrix 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)
- In-memory inference efficiency
- Aviator compiler / software
- Production readiness
- Ecosystem integrations
- 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
d-Matrix is graded here as runtimes, compilers & systems software. Criteria scores can move as more review volume and product checks are added.