decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) vs H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim)
As of Oct 14, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) 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.
H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim)
#17 Best current rank: #17 in Decision models, as of Oct 14, 2026.
| Category | decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) | H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) |
|---|---|---|
| Decision models | #4 · Near-SOTA · Decision models Margin 2.2727272727272663 JevBench · Oct 8, 2026 | #17 · #17 in Decision models Margin 7.575757575757578 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.
H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim): 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.
H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim): open weights.
Decision models: As of Oct 14, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) in Decision models at #4 against #17.
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) · H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim)
decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) or H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim)?
As of Oct 14, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) 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 H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) for Decision models?
As of Oct 14, 2026, decisio v0.8.0 on gemma-4-31B-it (frozen, FP8 on load, one prefill per state, self-hosted) leads H2O-Lightning-4B v1.1 (H2O.ai, Qwen3.5-4B fine-tune, stock vLLM + open shim) in Decision models at #4 against #17.