Lamb Labs
Chip company building Model Processing Units (MPUs) that hard-code an entire language model, weights included, into silicon to remove the memory-bandwidth bottleneck of GPUs, with a one-bit version of Qwen3.8 27B called Larry to chat with live and an FPGA prototype.
Product
Selected offerings — the AI products recorded here, not the vendor's full catalogue.
-
Model Processing Unit: a custom chip whose architecture and weights are hard-coded for one model, targeting 20,000+ tokens per second for a 27B-parameter model; not yet a chip on sale, with a one-bit model (Larry) to chat with live and an FPGA prototype as proof stages.
Similar products 20
No sourced competitor yet: no filing, vendor document or measured-market report read so far names a rival for this company, so none is asserted. The companies below are the closest by what their products do.
Closest alternatives are pairs a vendor page itself compares. Similar products are matched from what each product does; nothing in that group is asserted as a rivalry.
Sources
- Total funding
-
Pages read for this record state no funding total. Tried the home page and the company's YC profile.
- Valuation
-
Pages read for this record state no valuation. Tried the home page and the company's YC profile.
- Founded
-
Company disclosure · 19 Oct 2026
Own YC profile lists Founded: 2026; lamb-labs.com states no founding year. The pages name no legal entity.
- Public listing
-
No listing on the pages read; treated as private.
- Description
-
Official documentation · 19 Oct 2026
Home page: "Lamb Labs is building MPUs (Model Processing Units). We hardcode the entire model, including its weights, into silicon."
Open Lamb Labs in the directory
Something wrong here? Send a correction — quote this company id: lamb-labs.