Qwen 1.7B
A 1.7-billion-parameter small-scale language model in Alibaba's Qwen series, intended for edge deployment.
Timeline (last 90 days)
On Qwen 1.7B, the compiled regex Trie reduced median latency from 905.2 milliseconds to 567.5 milliseconds (a 37.3% speedup), but recall dropped sharply from 83.1% to 60.7%, a loss of 22.4 percentage points.
TypeSafe AI Jev achieves latency comparable to regex on Qwen 1.7B (576.2 milliseconds vs 567.5 milliseconds, statistically verified t=8.138, p=1.38×10⁻⁹) while restoring recall to 65.7%.
All Facts (2)
On Qwen 1.7B, the compiled regex Trie reduced median latency from 905.2 milliseconds to 567.5 milliseconds (a 37.3% speedup), but recall dropped sharply from 83.1% to 60.7%, a loss of 22.4 percentage points.
50%UnverifiedTypeSafe AI Jev achieves latency comparable to regex on Qwen 1.7B (576.2 milliseconds vs 567.5 milliseconds, statistically verified t=8.138, p=1.38×10⁻⁹) while restoring recall to 65.7%.
50%