Model comparison
DeepSeek V4.1 Flash vs Muse Spark 1.3
Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 7.6× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
DeepSeek
MetaSummary
- They share 33 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 1 category and Muse Spark 1.3 in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek V4.1 Flash leads 57.9 to 42.6.
- The biggest single-benchmark swing is FrontierMath Tier 4: 26.8% for DeepSeek V4.1 Flash and 46.3% for Muse Spark 1.3.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.3.
- Muse Spark 1.3 accepts more context: 1.05M tokens versus 1M.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4.1 Flash | Muse Spark 1.3 | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 52.8 | 54.8 |
| Released | 2026-09-09 | 2026-09-02 |
| Weights | Open | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 393K | 131K |
| Input $ / M tokens | $0.15 | $1.25 |
| Output $ / M tokens | $0.60 | $4.25 |
| Results tracked | 37 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 52.9 (#32), Muse Spark 1.3: 56.6 (#21)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena WebDev | 1619 | 1657 |
| SciCode | 51.9% | 59.7% |
| LMArena Coding | 1506 | 1514 |
| CursorBench | — | 41.6% |
| ALE-Bench | 1,092 | — |
Agentic & Tool Use Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 31.2 (#69), Muse Spark 1.3: 38.6 (#30)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| APEX-Agents | 39.5% | 57.8% |
| GDP.pdf | 19.8% | 27.6% |
Reasoning Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 50.2 (#36), Muse Spark 1.3: 54.0 (#27)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| NYT Connections (extended) | 89.6% | 85.1% |
| CritPt | 14.3% | 26% |
| LMArena Hard Prompts | 1483 | 1503 |
| Mystery Game Puzzles | 43% | 25% |
| DTBench | 89.9% | 96.5% |
| LMCA | 47% | 53.9% |
| Epoch Capabilities Index | 154.9 | 156.75 |
| Chess Puzzles | — | 38% |
| Surface Evolver Bench | 46.3% | — |
| Bench to the Future 3 | — | 0.14 |
Math Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 66.7 (#25), Muse Spark 1.3: 73.1 (#21)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 74.4% |
| FrontierMath Tier 4 | 26.8% | 46.3% |
| OTIS Mock AIME 2024-2025 | 98.3% | 99.2% |
| ProofBench | 54% | 58% |
| LMArena Math | 1477 | 1494 |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Muse Spark 1.3: 42.6 (#95)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Expert | 1506 | 1516 |
| GPQA Diamond | 89.8% | — |
Multimodal Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 39.1 (#61), Muse Spark 1.3: 43.7 (#22)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Vision | 1277 | 1309 |
| Furniture Assembly | 34.2% | — |
| LMArena Document | — | 1471 |
Multilingual Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 55.0 (#35), Muse Spark 1.3: 57.4 (#8)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Non-English | 1448 | 1481 |
| LMArena Chinese | 1497 | 1529 |
| LMArena French | 1452 | 1524 |
| LMArena German | 1484 | 1515 |
| LMArena Japanese | 1412 | 1474 |
| LMArena Korean | 1452 | 1501 |
| LMArena Russian | 1471 | 1490 |
| LMArena Spanish | 1459 | 1490 |
Instruction Following Too close to call
DeepSeek V4.1 Flash: 77.3 (#26), Muse Spark 1.3: 77.5 (#22)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1477 |
Long Context Too close to call
DeepSeek V4.1 Flash: 45.2 (#47), Muse Spark 1.3: 45.6 (#32)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Longer Query | 1475 | 1488 |
Writing & Preference Muse Spark 1.3 leads
DeepSeek V4.1 Flash: 65.4 (#48), Muse Spark 1.3: 73.6 (#9)
| Benchmark | DeepSeek V4.1 Flash | Muse Spark 1.3 |
|---|---|---|
| LMArena Text | 1462 | 1490 |
| LMArena Creative Writing | 1435 | 1455 |
| EQ-Bench Creative Writing | 1540 | 1906 |
| LMArena Multi-Turn | 1457 | 1482 |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Muse Spark 1.3?
Muse Spark 1.3 is the stronger model overall, scoring 54.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 7.6× less per token, which makes it the better buy when Muse Spark 1.3's lead doesn't matter for your workload.
Which is cheaper, DeepSeek V4.1 Flash or Muse Spark 1.3?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Muse Spark 1.3 lists at $1.25 and $4.25.
Is DeepSeek V4.1 Flash or Muse Spark 1.3 better for coding?
Muse Spark 1.3 scores higher on coding benchmarks: 56.6 versus 52.9 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.3 does, with 1.05M tokens against 1M.
How many benchmarks do DeepSeek V4.1 Flash and Muse Spark 1.3 share?
33 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Muse Spark 1.3 has 37.