decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) vs OpenJev (thinking, BF16)
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 OpenJev (thinking, BF16) in 1 of 1 shared categories: Decision models.
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.
OpenJev (thinking, BF16)
#14 Best current rank: #14 in Decision models, as of Oct 15, 2026.
| Category | decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) | OpenJev (thinking, BF16) |
|---|---|---|
| Decision models | #4 · Near-SOTA · Decision models Margin 2.2727272727272663 JevBench · Oct 8, 2026 | #14 · #14 in Decision models Margin 6.060606060606062 JevBench · Oct 8, 2026 |
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.
OpenJev (thinking, BF16): 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): open weights.
OpenJev (thinking, BF16): 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 OpenJev (thinking, BF16) in Decision models at #4 against #14.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) · OpenJev (thinking, BF16)
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) or OpenJev (thinking, BF16)?
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 OpenJev (thinking, BF16) in 1 of 1 shared categories: Decision models.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) or OpenJev (thinking, BF16) 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 OpenJev (thinking, BF16) in Decision models at #4 against #14.