Benchmark Almanac

Harvest Rush Training Evaluation

Qwen2.5-1.5B-Instruct 0.98 tops its leaderboard Llama 3.1-8B 0.81 leads the models we track
Not in model cards

Models decide whether a tractor should drive over or bypass animals and obstacles at a stated fuel cost. The default evaluation generates 300 held-out examples.

The September 2026 release reports supervised Qwen2.5-1.5B-Instruct at 0.98 mean reward and GRPO-trained Llama 3.1 8B at 0.81. These runs use the training environment’s holdout.

Made by
CompassionML
Released
September 2026
Kind
Deception and alignment
Level
Everyday
Scored
Checked automatically
Answers
Prediction
Tasks
All public
Measures
Propensity
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Llama 3.1-8B #203 on the capability index
Last model added
July 2024
Scores refreshed
November 21st, 2026
2
Entries on its own leaderboard, 1 model we track

Qwen2.5-1.5B-Instruct tops the leaderboard with 0.98 Llama 3.1-8B leads with 0.81

Best score per entry from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Llama 3.1-8B, at 0.81. This view shows only models we track, each at its best.

  1. 1 Qwen2.5-1.5B-Instruct SFT on the answer column, 2,000 examples 0.98
  2. 2 Llama 3.1-8B GRPO from scratch, 300 steps 0.81

Source: Official Harvest Rush Training Evaluation leaderboard

Source: Official Harvest Rush Training Evaluation leaderboard

#EntryScoreSettingDateSource
1 Qwen2.5-1.5B-Instruct 0.98 SFT on the answer column, 2,000 examples Official leaderboard
2 Llama 3.1-8B 0.81 GRPO from scratch, 300 steps Jul 2024 Official leaderboard
#ModelScoreSettingDateSource
1 Llama 3.1-8B 0.81 GRPO from scratch, 300 steps Jul 2024 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
0.65 0.70 0.75 0.80 0.85 0.65 0.70 0.75 0.80 0.85

Source: Official Harvest Rush Training Evaluation leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Safety and alignment

All 636 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior. Inkling 2 Mini 8.30%
  • Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions. Inkling 2 Small 4.70%
  • Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior. Inkling 2 Mini 9.80%

Data audit

7 of 8 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in July 2024, for Llama 3.1-8B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 1 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (2 entries).
  • Note Models scored Only 1 model we track, plus 1 other entry on its leaderboard.
  • Warning Newest score The newest came 2 years ago, in July 2024, for Llama 3.1-8B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Llama 3.1-8B (#203).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 2 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check