Also Parallella[1]. I have two of their 16 (+ 2 ARM cores) core demonstration units sitting in a box somewhere. They were fun, and I'm sad they didn't get much further - I think they did get a 64 core version done, but the longer term goal was a 4096 core grid. You could directly access the memory in each core, and there was a predictable cycle count to do so based on location in the grid, so you could write algorithms that'd run in lock-step if you wanted.

There's also XMOS[2] that appears to have morphed quite far from the original concept but is still focused on parallelism.

The challenge for both is that it's really hard to carve out a decent sized space alongside regular CPUs and GPUs. They can't compete with CPUs on single core performance, and so far can't compete with GPUs on concurrency. Both seemed to try to carve out space in embedded / low powered systems. XMOS seems to have somewhat succeeded (at least they're still alive), while Parallella/Adapteva failed.

[1] https://parallella.org/

[2] https://en.wikipedia.org/wiki/XMOS