Is there an uncensored K2 Horizon 3.7B?

K2 Horizon 3.7B from Institute of Foundation Models is ranked 36 of 100 distinct active open-weight models in this Artificial Analysis snapshot.

Research and availability status

This is a research guide, not an available Wild West API model listing. No uncensored or abliterated K2 Horizon 3.7B endpoint is confirmed here. An open-weight entry on AA does not establish that a derivative exists, works with a particular harness, or is hosted by Wild West API.

“Uncensored” describes reduced refusal behaviour or a service’s policy, not a standardized editing method. “Abliterated” refers to an intervention intended to reduce refusal-related behaviour through model-internal edits.

What rank 36 actually tells you

K2 Horizon 3.7B scores above 64 of the 100 selected entries: a strictly-lower-score percentile of 64% in this snapshot. No other selected entry shares its exact index score. This entry is not flagged as an estimated index in the source data.

It sits in the upper half of this selection. The relevant decision is whether its measured strengths fit the workload better than adjacent entries, not whether its rank is the highest available. Hosting cost and hardware requirements are outside this snapshot.

Base-model benchmark evidence

These are measurements attached to the evaluated base model, not an uncensored derivative. Missing values mean not reported here, never a zero score. Same-benchmark comparisons below use only entries with a reported measurement.

The highest numerical reported percentage for K2 Horizon 3.7B is GPQA at 69.2%; the lowest is Humanity’s Last Exam at 13.9%. These are numerical extremes, not evidence that one skill is stronger than another: benchmark percentages cannot be compared as if the tests had equal difficulty, scoring rules or task coverage. Compare GPQA with that same benchmark on another model, not with a different test.

Context window and a practical evaluation workload

At 524,288 tokens, a candidate workload is a large document collection or a repository with supporting specifications. A useful test would ask the model to trace a claim across distant files, then verify every citation. A large advertised window does not establish accurate retrieval throughout that window, and packing it fully can change latency and cost.

77 of the 100 entries have a smaller reported window than K2 Horizon 3.7B. As a planning example, allocating 75% of its reported window to source material gives 393,216 tokens, leaving 131,072 for instructions, conversation, tool messages and output combined. This is arithmetic for a test budget, not an endpoint limit or an assertion that output can use the entire remainder.

Reasoning settings and release comparisons

It is marked reasoning by AA. There is no within-model setting spread to quantify, so a claim that a lower budget preserves its score would be unsupported. A derivative should be tested in the actual setting you intend to use.

Other entries credited to Institute of Foundation Models provide a publisher-level comparison, not proof of a shared architecture or training recipe. K2 Horizon 7B ranks 27, scores 20.60 and reports 524,288 context tokens; K2 Think V2 ranks 52, scores 11.50 and reports 262,144 context tokens; K2 Horizon MoVA 36B A4B ranks 17, scores 25.30 and reports 524,288 context tokens. Compare exact releases and evaluation settings rather than carrying a family’s reputation over to this model.

Nearby ranked alternatives

The closest ranks give a bounded shortlist for evaluating K2 Horizon 3.7B. A small index gap is not a statistical significance claim; compare task outcomes and deployment conditions directly.

How to evaluate an uncensored or abliterated candidate

For K2 Horizon 3.7B, first locate the exact base release and any claimed derivative’s model card. Verify access, redistribution and usage terms at their original sources: AA’s open-weight classification does not assert an open-source license. Record the base revision, derivative revision, runtime and reasoning setting so your comparison can be reproduced.

Weight-based abliteration generally studies differences between activations on refusal-inducing and ordinary prompts, identifies candidate refusal-related directions, and intervenes on model internals. It is not equivalent to changing a system prompt or removing an API filter. Whether a particular intervention is supported depends on the model and implementation; this guide supplies no architecture-specific recipe or claim that the method has been validated on K2 Horizon 3.7B.

Run the base and candidate on the same permitted task set before and after changes. Measure refusal behaviour separately from factual errors, instruction following, coding correctness and tool execution. Use GPQA as one reference for task coverage, while keeping its published percentage separate from your own results. Track any regression and repeat with the exact context length and setting your application needs. A lower refusal rate is not evidence that all other behaviour survived unchanged.

Sources and scope

Snapshot retrieved 2026-10-06T06:17:07.812043+00:00. AA’s K2 Horizon 3.7B model entry supplies the evaluated setting and measurements; the Artificial Analysis leaderboard is the population source. Top 100 distinct active open-weight model names by Artificial Analysis Intelligence Index. Reasoning-budget parentheticals are removed for grouping; the strongest scored setting represents each group. Open weights is AA's classification, not an open-source-license assertion. Scores describe the evaluated base models, not uncensored derivatives. Rank is within this selected active open-weight population, not the complete leaderboard.

For the research behind refusal-direction interventions, see Arditi et al., Refusal in Language Models Is Mediated by a Single Direction. This general research source does not establish successful abliteration of this release. Continue with the abliteration explanation or the current Wild West API line; only the latter lists products for sale.

Questions about K2 Horizon 3.7B