Public KG record
qwenGPU
Recorded overview
Compiles a SIMT GPU into Qwen3.5-2B's weights, running real CUDA inside its forw
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: reference interpreter matched NVIDIA Tesla T4 on 49 of 49 kernels
- Benchmarks: Qwen-written CUDA passed 4 of 5
- Benchmarks: reference interpreter matched an NVIDIA Tesla T4 on 49 of 49 kernels
- Benchmarks: answered 20 of 24 sampled chat questions with GPU-computed numbers
- Benchmarks: wrote a correct kernel first try on 4 of 5 questions
- Capabilities: supports integer arithmetic, predicates, divergent branches, loops
- Capabilities: supports shared memory, atomicAdd, __syncthreads(), warp shuffles
- Capabilities: supports integer arithmetic, predicates, divergent branches and loops
Evidence coverage
The KG currently records 10 known information dimensions and has 7 undocumented dimensions, linked to 1 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- Percepta — competes with
- Qwen3.5-2B — based on
- Percepta — competes with
- Qwenputer — competes with
- NVIDIA Tesla T4 — evaluated on