Diffusion language model with tunable reasoning and parallel tool calls. Listed prices are standard rates, excluding the launch discount. Independent benchmark coverage is pending.
| Benchmark / source | Score |
|---|---|
GPQA Diamond Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 79% |
Tau3Bench Telecom Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 96% |
IFBench Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 77% |
AA-LCR Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 68% |
SciCode Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 38% |
TerminalBench (version unspecified) Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 37% |
DSQA Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified; 10 tool calls | 34% |
Omniscience Non-Hallucination Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 33% |
Omniscience Accuracy Provider reportInception launch chart; effort unspecified; TerminalBench version unspecified | 22% |