RankingEffort benchmarks
Muse Glimmer
16 measured results · 1 effort settings · 16 benchmark/harness combinations
Change model
Capability profile
One configuration, seven capabilities
- AMuse Glimmer · High2 of 7 capabilities supported
Gaps are unknown, not zero. Hollow points are preliminary.
Agentic
A20.7SupportedHard reasoning
A20.6SupportedCoding
A21.2PreliminaryHuman pref
UnknownKnowledge
A29.1PreliminaryMultimodal
A21.1PreliminaryLong context
A67.2Preliminary
Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.
Capability score evidence
Muse Glimmer · High: 13.4 overall comparison score. Meets the overall evidence requirements.
Which sources carry the weight?
- Artificial Analysis · 100.0% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
- Humanity’s Last Exam · 7.7% direct comparison weight
- GPQA · 7.7% direct comparison weight
- Critpt · 7.7% direct comparison weight
- Mmmu Pro · 7.6% direct comparison weight
- GDPval · 7.1% direct comparison weight
- Other families (10) · 62.1% direct comparison weight
These shares describe direct comparison weight, not fractions of the final score or confidence. Source balancing does not give every family equal weight or remove correlations between overlapping sources.
How sensitive is this score?
The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-11-22.
Both ranks use the same surviving configurations for each removal. The cohort starts with 602 configurations supported in the original audit, including separate effort settings. These are diagnostic positions, not the default leaderboard ranks. Eligibility is not rechecked after removal.
| Removed source | Rank before → after | Rank movement | Cohort remaining / lost support | Score after (change) |
|---|---|---|---|---|
| Artificial Analysis | Comparison support lost; no rank | 230 remaining / 372 lost | Unavailable | |
| ARC Prize | 231 → 229 | Up 2 | 602 remaining / 0 lost | 16.4 |
| Datacurve | 231 → 231 | No movement | 602 remaining / 0 lost | 10.9 |
| LiveCodeBench | 231 → 232 | Down 1 | 602 remaining / 0 lost | 13.3 |
| Scale | 231 → 232 | Down 1 | 602 remaining / 0 lost | 14.7 |
| Terminal-Bench | 231 → 233 | Down 2 | 602 remaining / 0 lost | 13.4 |
| Cognition | 231 → 234 | Down 3 | 602 remaining / 0 lost | 12.6 |
| LiveBench | 230 → 227 | Up 3 | 601 remaining / 1 lost | 14.0 |
| SWE-rebench | 231 → 231 | No movement | 602 remaining / 0 lost | 13.6 |
| Epoch AI | 226 → 231 | Down 5 | 588 remaining / 14 lost | 13.8 |
| Vals AI | 231 → 233 | Down 2 | 601 remaining / 1 lost | 12.9 |
| Provider reports · Anthropic | 231 → 231 | No movement | 602 remaining / 0 lost | 13.4 |
| Provider reports · Moonshot AI | 231 → 233 | Down 2 | 602 remaining / 0 lost | 13.1 |
| Provider reports · Meta | 231 → 231 | No movement | 602 remaining / 0 lost | 13.4 |
| Provider reports · Mistral | 231 → 231 | No movement | 602 remaining / 0 lost | 13.4 |
| Provider reports · OpenAI | 231 → 231 | No movement | 602 remaining / 0 lost | 13.5 |
| All provider reports | 231 → 233 | Down 2 | 602 remaining / 0 lost | 13.1 |
Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.
Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit
Every benchmark family
| Removed evidence | Recomputed score | Change |
|---|---|---|
| arc-agi | 15.9 | +2.6 |
| deepswe | 10.9 | -2.4 |
| aa-lcr | 11.8 | -1.6 |
| terminal-bench-science | 14.4 | +1.0 |
| briefcase | 14.3 | +1.0 |
| mmmu-pro | 14.2 | +0.8 |
| frontiercode | 12.6 | -0.7 |
| hle | 14.1 | +0.7 |
| enterprise-ops | 12.7 | -0.7 |
| tau-bench | 12.7 | -0.7 |
| gdpval | 14.0 | +0.7 |
| omniscience | 12.8 | -0.6 |
| simpleqa | 13.9 | +0.5 |
| epoch-game-puzzles | 12.9 | -0.5 |
| ifbench | 13.0 | -0.4 |
| analyst-agent | 13.0 | -0.3 |
| mmlu-pro | 13.7 | +0.3 |
| automationbench | 13.7 | +0.3 |
| livebench-instructions | 13.6 | +0.3 |
| terminal-bench | 13.6 | +0.3 |
| enigma-eval | 13.6 | +0.3 |
| swe-rebench | 13.6 | +0.2 |
| multi-swe-bench | 13.1 | -0.2 |
| arc-agi-3 | 13.2 | -0.2 |
| apex-agents | 13.6 | +0.2 |
| livebench-language | 13.6 | +0.2 |
| critpt | 13.5 | +0.2 |
| livebench-coding | 13.2 | -0.1 |
| swe-atlas-qna | 13.2 | -0.1 |
| vibe-code | 13.3 | -0.1 |
| aime | 13.3 | -0.1 |
| gpqa | 13.5 | +0.1 |
| vals-code-migration | 13.3 | -0.1 |
| livebench-reasoning | 13.3 | -0.1 |
| vals-legal-research | 13.3 | -0.1 |
| itbench | 13.3 | -0.1 |
| vals-finance-agent | 13.3 | -0.1 |
| swe-atlas-test-writing | 13.3 | -0.1 |
| vals-tax-agent | 13.3 | -0.1 |
| gmmlu | 13.4 | +0.1 |
| vals-excel-modeling | 13.3 | -0.0 |
| swe-bench-pro | 13.4 | +0.0 |
| math500 | 13.4 | +0.0 |
| frontiermath | 13.3 | -0.0 |
| livecodebench | 13.4 | +0.0 |
| gdp-pdf | 13.4 | +0.0 |
| swe-atlas-refactoring | 13.4 | -0.0 |
| livebench-data | 13.4 | +0.0 |
| livebench-math | 13.4 | +0.0 |
| harvey | 13.4 | +0.0 |
What evidence is missing from the fit?
Across all collected settings for this model: 16 observations, 16 contributing, 0 matched without graph weight and 0 excluded. Counts do not establish rank eligibility.
Benchmark results
Reviewed retained observations are in use. October 2026 Real-SWE board has no 95% confidence whisker, so these scores stay at the 2026-10-18 retrieval. Original source retrieval 2026-10-18; source updated 2026-10-18.
| Benchmark / source | High |
|---|---|
AA-AnalystAgentBenchmark detailsShare of tasks solved on all 5 of 5 attempts · 80 tasks, 5 attempts per task · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 20% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
AA-Briefcase EloBenchmark detailsAA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-11-22. | 478.09 Elo Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
AA-LCR v1.1Benchmark detailsIndependently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 83.33% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
AutomationBench-AABenchmark detailsShare of task objectives completed with no guardrail violations · Higher is better · Benchmark developed by Zapier · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 6.83% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
CritPtBenchmark detailsBenchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 2.57% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
EnterpriseOps-Gym-AABenchmark detailsOverall task success rate across all eight enterprise domains (oracle tool mode) · Benchmark developed by ServiceNow Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 34.74% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
GDP.pdf: All-pass RateBenchmark detailsShare of attempts where every atomic criterion passed · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 10% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
GDPval-AA v2.1 LeaderboardBenchmark detailsElo rating for performance on real-world work tasks · Anchored to DeepSeek V4.1 Flash (max) at 1600 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-11-22. | 791.7 Elo Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
GPQA DiamondBenchmark detailsIndependently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 83.54% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
Humanity's Last ExamBenchmark detailsIndependently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 21.96% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
MMMU-ProBenchmark detailsIndependently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 74.34% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
AA-Omniscience AccuracyBenchmark detailsAA-Omniscience v1.1 Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 27.97% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
𝜏³-BankingBenchmark detailsBenchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 23.51% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
Terminal-Bench-Science 0.1Benchmark detailsShare of tasks passed (pass@1), averaged over three repeats · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 0% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
Terminal-Bench v2.1Benchmark detailsBenchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 51.69% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
Terminal-Bench v4.0Benchmark detailsBenchmark developed by the Laude Institute, Stanford, and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-11-22. | 0.51% Reported settingsMuse Glimmer (High); Model source: https://artificialanalysis.ai/models/muse-glimmer |
What these comparisons mean
Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.
These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.
Download all effort benchmark results · 10,184 results across 381 models · Collected 2026-11-22