Model comparison
Muse Spark vs Trinity-Large-Preview
Head-to-head evidence from 14 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Muse Spark #13 (Supported); Trinity-Large-Preview #79 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Muse Spark and Trinity-Large-Preview share 14 comparable benchmark results. 0 of 8 categories are comparable. 25 results are unique to Muse Spark; 4 to Trinity-Large-Preview.
Updated July 21, 2026- Shared results
- 14
- Muse Spark only
- 25
- Trinity-Large-Preview only
- 4
- Comparable categories
- 0 / 8
Benchmark data for Muse Spark and Trinity-Large-Preview is coming soon on BenchLM.
Confidence note. This is a partial-evidence comparison with 14 shared benchmark results across 5 evidence categories; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.
Trinity-Large-Preview has the larger context window at 512K, compared with 262K for Muse Spark.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Muse Spark | Δ | Trinity-Large-Preview |
|---|---|---|---|
| Agentic | Muse Spark59.0 | MarginNo overlap | Trinity-Large-PreviewNot measured |
| Coding | Muse Spark67.8 | MarginNo overlap | Trinity-Large-PreviewNot measured |
| Reasoning | Muse Spark42.5 | MarginNo overlap | Trinity-Large-PreviewNot measured |
| Knowledge | Muse Spark50.4 | MarginNo overlap | Trinity-Large-PreviewNot measured |
| Math | Muse Spark32.9 | MarginNo overlap | Trinity-Large-PreviewNot measured |
| Multimodal | Muse Spark82.5 | MarginNo overlap | Trinity-Large-PreviewNot measured |
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Muse Spark | Trinity-Large-Preview | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Muse SparkNot available | Trinity-Large-Preview$0.25 input / $1 output | A complete price comparison is not available. |
| Generation speedtokens per second | Muse SparkNot available | Trinity-Large-PreviewNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Muse SparkNot available | Trinity-Large-PreviewNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Muse Spark262K | Trinity-Large-Preview512K | Trinity-Large-Preview lists the larger context window. |
Benchmark Deep Dive
Agentic8 benchmarks
| Benchmark | Muse Spark | Trinity-Large-Preview | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 59% | — | Not comparable |
| τ²-bench resultsSource | 91.5% | 90.1% | Muse Spark leads |
| DeepSearchQASource | 74.8% | — | Not comparable |
| CyberGymSource | 43.5% | — | Not comparable |
| Claw-EvalSource | 63.8% | — | Not comparable |
| AA Agentic IndexSource | 28.7% | — | Not comparable |
| GDPval-AASource | 32.2% | 2.7% | Muse Spark leads |
| GDPval-AASource | 1144 | 554 | Muse Spark leads |
Coding6 benchmarks
| Benchmark | Muse Spark | Trinity-Large-Preview | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 77.4% | — | Not comparable |
| SWE-bench ProSource | 52.4% | — | Not comparable |
| LiveCodeBench ProSource | 80.0% | — | Not comparable |
| Vibe Code BenchSource | 19.67% | — | Not comparable |
| AA Coding IndexSource | 58.6% | — | Not comparable |
| AA-SciCodeSource | 51.5% | 36.1% | Muse Spark leads |
Reasoning3 benchmarks
Knowledge13 benchmarks
| Benchmark | Muse Spark | Trinity-Large-Preview | Result |
|---|---|---|---|
| GPQA-DSource | 89.5% | 63.3% | Muse Spark leads |
| HLESource | 50.4% | — | Not comparable |
| HLE w/o toolsSource | 42.8% | — | Not comparable |
| HealthBench HardSource | 42.8% | — | Not comparable |
| MedXpertQA (Text)Source | 52.6% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 43.1% | 24.5% | Muse Spark leads |
| AA-GPQA DiamondSource | 88.4% | 75.2% | Muse Spark leads |
| AA-HLESource | 39.9% | 14.7% | Muse Spark leads |
| AA-Omniscience IndexSource | 4.1% | -44.2% | Muse Spark leads |
| AA-Omniscience AccuracySource | 44.6% | 22.8% | Muse Spark leads |
| AA-Omniscience Hallucination RateSource | 73.2% | 86.6% | Muse Spark leads |
| MMLUSource | — | 87.2% | Not comparable |
| MMLU-Pro (Arcee)Source | — | 75.2% | Not comparable |
Math3 benchmarks
Multimodal9 benchmarks
| Benchmark | Muse Spark | Trinity-Large-Preview | Result |
|---|---|---|---|
| CharXivSource | 86.4% | — | Not comparable |
| MMMU-ProSource | 80.4% | — | Not comparable |
| ERQASource | 64.7% | — | Not comparable |
| SimpleVQASource | 71.3% | — | Not comparable |
| ScreenSpot ProSource | 84.1% | — | Not comparable |
| ZeroBenchSource | 33.0% | — | Not comparable |
| MedXpertQA (MM)Source | 78.4% | — | Not comparable |
| AA-MMMU-ProSource | 80.5% | — | Not comparable |
| Design Arena WebsiteSource | — | 1165 | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Muse Spark | Trinity-Large-Preview | Result |
|---|---|---|---|
| AA-IFBenchSource | 75.9% | 56.3% | Muse Spark leads |
Frequently Asked Questions (3)
Can I compare Muse Spark and Trinity-Large-Preview on BenchLM yet?
Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.
Why does this comparison show “coming soon”?
BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.
What data is available for Muse Spark and Trinity-Large-Preview today?
Trinity-Large-Preview: $0.25 input / $1.00 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.
Related Comparisons
Explore More
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.