MatX
Last verified 2026-06-02. JSON
The funding figures on this page are attributed, not sourced: a publisher is on record but no document URL is. How we mark provenance.
| Total raised | $105.0Mattributed |
|---|---|
| Latest round | series-a, $80.0M, 2024-11-22attributed |
| Founded | 2022 |
| Headquarters | USA |
| Industry | infra |
| Stage | series-a |
| Website | matx.com |
MatX designs accelerator chips, marketed as MatX One, purpose-built for training and serving large language models, with weights in SRAM and HBM for KV cache. It targets frontier AI labs running training, RL, and inference at hundred-thousand-chip scale. The architecture pushes the highest FLOPS per square millimeter and clears 2,000 output tokens per second on huge MoE models.
Funding rounds
| Date | Round | Amount | Lead |
|---|---|---|---|
| 2024-11-22 | series-a | $80.0M | undisclosed |
Detected tech stack
hsts, vercel
Provenance
| Fact | Status | Basis |
|---|---|---|
| Series A, $80.0M | attributed | techcrunch named as publisher. No document URL on file. |
| Total raised $105.0M | attributed | techcrunch named as publisher. No document URL on file. |
| Company existence | self-reported | matx.com, read 2026-06-02 |
| Description | self-reported | matx.com, read 2026-06-02 |
Nothing on this page is sourced yet. A sourced fact links to a document that states it.
Change history
No change recorded since records began on 2026-09-24.