Model comparison
GPT-5 Mini vs Muse Spark
Muse Spark is the stronger model overall, scoring 50.6 to 41.8 on the Noometry Index.
Last verified . 26 shared benchmarks.
MetaSummary
- They share 26 benchmarks with published results for both. GPT-5 Mini scores higher in 1 category and Muse Spark in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Muse Spark leads 65.7 to 45.6.
- The biggest single-benchmark swing is Humanity's Last Exam: 19.4% for GPT-5 Mini and 40.6% for Muse Spark.
Side by side
| GPT-5 Mini | Muse Spark | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 41.8 | 50.6 |
| Released | 2025-08-07 | 2026-04-08 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $2 | — |
| Results tracked | 60 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark leads
GPT-5 Mini: 40.1 (#146), Muse Spark: 46.2 (#69)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| SciCode | 39.2% | 51.5% |
| LMArena Coding | 1406 | 1481 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| WeirdML | 52.7% | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Not comparable
GPT-5 Mini: 31.1 (#70), Muse Spark: —
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning Muse Spark leads
GPT-5 Mini: 23.9 (#168), Muse Spark: 35.9 (#67)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| CritPt | 0% | 11.3% |
| LMArena Hard Prompts | 1380 | 1474 |
| Epoch Capabilities Index | 145.52 | 152.04 |
| ARC-AGI-2 | 4.4% | — |
| Kagi LLM Benchmark | 70.3% | — |
| ARC-AGI-1 | 54.3% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LMCA | 34.2% | — |
| ForecastBench | 61 | — |
Math Muse Spark leads
GPT-5 Mini: 46.7 (#69), Muse Spark: 47.8 (#66)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 88.9% |
| ProofBench | 9% | 17% |
| LMArena Math | 1378 | 1455 |
| FrontierMath (Feb 2025 set) | 27.2% | 39% |
| FrontierMath Tier 4 (v1) | 6.3% | 14.6% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| Omni-MATH | 72.2% | — |
| MATH Level 5 | 97.8% | — |
Knowledge Muse Spark leads
GPT-5 Mini: 45.6 (#86), Muse Spark: 65.7 (#13)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| GPQA Diamond | 75% | 89.8% |
| Humanity's Last Exam | 19.4% | 40.6% |
| LMArena Expert | 1379 | 1457 |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| Vectara Hallucination Rate | 12.9% | — |
| GPQA (HELM) | 75.6% | — |
Multimodal Muse Spark leads
GPT-5 Mini: 35.6 (#85), Muse Spark: 43.4 (#24)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| LMArena Vision | 1202 | 1306 |
| VPCT | 40.2% | — |
| LMArena Document | — | 1444 |
Multilingual Muse Spark leads
GPT-5 Mini: 48.9 (#137), Muse Spark: 56.1 (#24)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| LMArena Non-English | 1363 | 1464 |
| LMArena Chinese | 1385 | 1509 |
| LMArena French | 1386 | 1497 |
| LMArena German | 1366 | 1497 |
| LMArena Korean | 1308 | 1459 |
| LMArena Russian | 1362 | 1466 |
| LMArena Spanish | 1355 | 1472 |
| LMArena Japanese | 1341 | — |
Instruction Following Too close to call
GPT-5 Mini: 76.2 (#46), Muse Spark: 75.9 (#51)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| LMArena Instruction Following | 1357 | 1442 |
| IFEval | 92.7% | — |
Long Context Muse Spark leads
GPT-5 Mini: 41.9 (#132), Muse Spark: 44.4 (#69)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| LMArena Longer Query | 1355 | 1451 |
| Fiction.LiveBench | 69.4% | — |
Writing & Preference Muse Spark leads
GPT-5 Mini: 55.2 (#148), Muse Spark: 66.0 (#39)
| Benchmark | GPT-5 Mini | Muse Spark |
|---|---|---|
| LMArena Text | 1373 | 1474 |
| LMArena Creative Writing | 1325 | 1459 |
| LMArena Multi-Turn | 1363 | 1477 |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
Frequently asked questions
Is GPT-5 Mini better than Muse Spark?
Muse Spark is the stronger model overall, scoring 50.6 to 41.8 on the Noometry Index.
Is GPT-5 Mini or Muse Spark better for coding?
Muse Spark scores higher on coding benchmarks: 46.2 versus 40.1 in the Noometry coding category.
How many benchmarks do GPT-5 Mini and Muse Spark share?
26 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Muse Spark has 27.