Benchmark Almanac

BioTIER v1 Refuse

Muse Spark 98% leads since June 2026
Only in system cards

Models receive 501 dangerous biology prompts on select agents, dual-use research and catastrophe avoidance. The score is the share correctly refused.

The version Meta reports for its Muse models. It includes select agent prompts, which the 398-prompt BioTIER Refuse track leaves out, so the scores do not compare.

Made by
SecureBio
Released
June 2026
Kind
Refusing harmful requests
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
Kept private
Measures
Propensity
Behavior
Harmful use
Language
English
In and out
Text → Text
Interaction
Single turn
Run by
A host
Strongest model tested
Muse Spark 1.3 #27 on the capability index
Last in a system card
October 2026
Last model added
October 2026
3
Models scored in lab cards
3
System cards report it, from 1 lab

Muse Spark leads with 98% among models released by September 2026

Best score per model as labs report them in their own cards, drawn striped. Labs pick their own settings, so these compare less cleanly than independent runs.

The newest model with a score here came out in September 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Muse Spark 98%
  2. 2 Muse Spark 1.1 97.7%
  3. 3 Muse Spark 1.3 97.4%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 Muse Spark 98% Jun 2026 Lab model card
2 Muse Spark 1.1 97.7% Jul 2026 Lab model card
3 Muse Spark 1.3 97.4% Oct 2026 Lab model card

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Meta
Lab claims Passed 85% in Jun 2026, by Muse Spark

Source: Labs’ model cards

Meta’s own number for Muse Spark passed 85% in June 2026

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
Reported by the lab in its model card
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Muse Spark 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Muse Spark

Source: Labs’ model cards

Labs report it only in system cards

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

3 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Oct 23, 2026 Muse Spark 1.3 system card 97.4 BioTIER v1 (English) refusal rate, Muse Spark 1.3 (API) with system mitigations (Table 2, Figure 5). Multilingual 96.9, adversarial 95.3 Bio and chem risk
Jul 9, 2026 Muse Spark 1.1 system card 97.7 Refusals: BioTIER v1 refusal rate (Table 2, Figure 6) Bio and chem risk
Jun 12, 2026 Muse Spark system card 98 Refusals: BioTIER refusal rate, deployed system (Table 2, Figure 9) Bio and chem risk

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

All 4 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 1 pass 4 notes
  • Note Independent runs Only labs’ own numbers, as expected for an in-house eval. Scores from system cards (3 scores).
  • Note Models scored Only 3 models we track.
  • Pass Newest score The newest is from October 2026, for Muse Spark 1.3.
  • Note Frontier models None of the top 10 has a score, which is fine for an in-house eval. The strongest tested is Muse Spark 1.3 (#27).
  • Note Headroom Saturated. The best is 98%, past 85% since June 2026.
Leaderboard import 1 note
  • Note Importer No public leaderboard to read.
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