Model comparison
DeepSeek-V3.1-Terminus vs Muse Spark
Muse Spark is the stronger model overall, scoring 50.6 to 43.1 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
MetaSummary
- They share 12 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and Muse Spark in 7 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark leads 35.9 to 26.4.
- The biggest single-benchmark swing is SciCode: 40.6% for DeepSeek-V3.1-Terminus and 51.5% for Muse Spark.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | Muse Spark | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 43.1 | 50.6 |
| Released | 2025-09-22 | 2026-04-08 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Muse Spark: 46.2 (#69)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| SciCode | 40.6% | 51.5% |
| LMArena Coding | 1426 | 1481 |
| ALE-Bench | 745.17 | — |
Reasoning Muse Spark leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Muse Spark: 35.9 (#67)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| CritPt | 1.7% | 11.3% |
| LMArena Hard Prompts | 1426 | 1474 |
| Kagi LLM Benchmark | 57.4% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 152.04 |
Math Muse Spark leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Muse Spark: 47.8 (#66)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Math | 1402 | 1455 |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 17% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Muse Spark: 65.7 (#13)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| GPQA Diamond | — | 89.8% |
| Humanity's Last Exam | — | 40.6% |
| LMArena Expert | — | 1457 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Muse Spark: 43.4 (#24)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Vision | — | 1306 |
| LMArena Document | — | 1444 |
Multilingual Muse Spark leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Muse Spark: 56.1 (#24)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Non-English | 1407 | 1464 |
| LMArena Russian | 1436 | 1466 |
| LMArena Chinese | — | 1509 |
| LMArena French | — | 1497 |
| LMArena German | — | 1497 |
| LMArena Korean | — | 1459 |
| LMArena Spanish | — | 1472 |
Instruction Following Muse Spark leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Muse Spark: 75.9 (#51)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Instruction Following | 1404 | 1442 |
Long Context Too close to call
DeepSeek-V3.1-Terminus: 43.4 (#97), Muse Spark: 44.4 (#69)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Longer Query | 1421 | 1451 |
Writing & Preference Muse Spark leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Muse Spark: 66.0 (#39)
| Benchmark | DeepSeek-V3.1-Terminus | Muse Spark |
|---|---|---|
| LMArena Text | 1419 | 1474 |
| LMArena Creative Writing | 1403 | 1459 |
| LMArena Multi-Turn | 1411 | 1477 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Muse Spark?
Muse Spark is the stronger model overall, scoring 50.6 to 43.1 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Muse Spark better for coding?
Muse Spark scores higher on coding benchmarks: 46.2 versus 42.0 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Muse Spark share?
12 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Muse Spark has 27.