Bobcat 1.3 (31B) vs decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted)
As of Oct 15, 2026, Bobcat 1.3 (31B) leads decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) in 0 of 1 shared categories.
Bobcat 1.3 (31B)
Yes. Yes on Decision models (same-GPU latency). Not the overall leader.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted)
Yes. Yes on Decision models (price per 1,000 decisions). Not the overall leader.
| Category | Bobcat 1.3 (31B) | decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) |
|---|---|---|
| Decision models | #5 · Near-SOTA · Decision models Margin 2.424242424242422 Jev Decision Index · Oct 14, 2026 | #4 · Near-SOTA · Decision models Margin 2.2727272727272663 JevBench · Oct 8, 2026 |
Bobcat 1.3 (31B): price per 1M tokens is unset in the catalog.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted): price per 1M tokens is unset in the catalog.
Bobcat 1.3 (31B): open weights.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted): open weights.
Decision models: As of Oct 15, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads Bobcat 1.3 (31B) in Decision models at #4 against #5.
Bobcat 1.3 (31B) · decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted)
Bobcat 1.3 (31B) or decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted)?
As of Oct 15, 2026, Bobcat 1.3 (31B) leads decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) in 0 of 1 shared categories.
Bobcat 1.3 (31B) or decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) for Decision models?
As of Oct 15, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads Bobcat 1.3 (31B) in Decision models at #4 against #5.