Gemma 4 31B Ortenzya Creative Wordsmith (Thinking)
This community fine-tune variant of Gemma 4 31B targets creative writing applications and foregoes most of the base model’s safety filters, with additional fine-tuning for a more natural writing style. The dense Open Weights model with 30.7 billion parameters and 256,000 tokens of context runs locally as an NVFP4 variant with low memory requirements. Apache 2.0 license inherited from the base model.
- Open Weights
- Workstation
- vLLM
- Text
- Vision
- Uncensored
- Instruction-Tuned
- Unusable
Sovereign Risk: MEDIUM Google DeepMind is a US company (CLOUD Act exposure applies primarily to cloud/API usage, not local deployment). The base weights are licensed under Apache-2.0. Lineage: google/gemma-4-31B → google/gemma-4-31B-it → llmfan46/gemma-4-31B-it-uncensored-heretic (abliteration via Heretic v1.2.0, ARA method) → llmfan46/…/Ortenzya-Creative-Wordsmith (fine-tune via Unsloth Studio) → NVFP4 quantization by the same author. The fine-tune author llmfan46 is a solo contributor with no documented jurisdiction (HF profile lists no country). Relevant risk factor: the model was deliberately abliterated (91% fewer refusals, 9/100 vs. 99/100 for the original), meaning the base model’s safety guardrails have been intentionally removed — when running purely locally without cloud connectivity, the risk is technical/content-related, not a data privacy concern.
Key metrics
Score · Latency · Cost · Quality
- Total Score Silver
- 72.56
- Routine
- 43.42
- Reasoning
- 29.14
- LLM Judge Avg
- 3.74 / 5
- 100 Coverage
- Avg Task Duration
- 192.11s
- Unusable
- Token Rate
- 6.65tok/s
- Output Rate
- P95 Latency
- 444.69s
- Top 5 %
- Total Tokens
- 119400
- Output Volume
- Cost per 1K
- $0
- USD / 1K Requests
- Benchmark Cost
- $0
- Total · 119400 tok
Benchmark modules
10 modules · weighted · vs. model median & top performer
Token efficiency & latency
Consumption per module vs. model median
Token consumption per module
Performance profile
Key metrics – legend
Score
- Total Score
- Overall score derived from Routine and Reasoning scores. Basis for the fleet rank and tier assignment (Gold / Silver / Bronze).
- Routine
- Score for standard tasks: code generation, text transformation, structured outputs. Weighted by task complexity.
- Reasoning
- Score for analytical and logical tasks: multi-step problem solving, mathematical reasoning, structured argumentation.
- Rank
- Position in the fleet comparison, sorted by Total Score. Refers to the snapshot at the time of testing.
LLM Judge
- LLM Judge Ø
- Average quality rating by an independent judge model on a scale of 1–5.
- Coverage
- Share of tasks for which a judge rating is available. 100% means all responses were evaluated.
Latency & speed
- Avg. Task Duration
- Average time per task from the first to the last token of the response, measured across all benchmark tasks.
- Token Rate
- Output speed in tokens per second. Relevant for interactive applications and real-time streaming.
- P95 Latency
- 95th percentile of response times — shows outliers. Lower values mean fewer long wait times in edge cases.
Cost
- Cost per 1K
- API price per 1,000 output tokens in US dollars. Basis for cost calculation in production use.
- Benchmark Cost
- Total cost for the complete benchmark run of this model.