Compare / head-to-head
Muse Spark 1.2vsMuse Spark 1.3
Muse Spark 1.3 leads 14 of 16 shared benchmarks. Both list the same combined token price. Both offer a 1.0M-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
2 – 14
Muse Spark 1.3 leads
Cheaper per token
Tie
list price, input + output
Larger context
Tie
both 1.0M tokens
Providers
Meta
same provider
Muse Spark 1.2
Meta · released 2026-08-05
textvisionvideoaudio
Muse Spark 1.3
Meta · released 2026-09-02
textvisionvideo
Quality
Benchmark matrix
16 shared · 7 only Muse Spark 1.2 · 13 only Muse Spark 1.3| Benchmark | Muse Spark 1.2 | Muse Spark 1.3 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 16 | ||||
| LMArena Elo | 1493.5 ↗ | 1494.3 ↗ | -0.8 | Muse Spark 1.3 |
| APEX-Agents (Mercor) | 36.4% ↗ | 57.8% ↗ | -21.4 pt | Muse Spark 1.3 |
| DeepSWE v1.1 | 59.3% ↗ | 75.4% ↗ | -16.1 pt | Muse Spark 1.3 |
| LiveBench Agentic Coding (LiveBench) | 57.6% ↗ | 64.1% ↗ | -6.5 pt | Muse Spark 1.3 |
| LiveBench Coding (LiveBench) | 77.5% ↗ | 81.1% ↗ | -3.6 pt | Muse Spark 1.3 |
| LiveBench Data Analysis (LiveBench) | 76.5% ↗ | 79.6% ↗ | -3.1 pt | Muse Spark 1.3 |
| LiveBench Instruction Following (LiveBench) | 74.3% ↗ | 78% ↗ | -3.7 pt | Muse Spark 1.3 |
| LiveBench Language (LiveBench) | 78.6% ↗ | 82.8% ↗ | -4.2 pt | Muse Spark 1.3 |
| LiveBench Mathematics (LiveBench) | 91.2% ↗ | 95.9% ↗ | -4.7 pt | Muse Spark 1.3 |
| LiveBench Reasoning (LiveBench) | 90% ↗ | 89.7% ↗ | +0.3 pt | Muse Spark 1.2 |
| LMArena Agent (LMArena) | -0.0327 ↗ | 0.04 ↗ | -0.1 | Muse Spark 1.3 |
| LMArena Vision (LMArena) | 1292.8 ↗ | 1290.2 ↗ | +2.6 | Muse Spark 1.2 |
| LMArena WebDev (LMArena) | 1531.8 ↗ | 1656.6 ↗ | -124.8 | Muse Spark 1.3 |
| SimpleBench (SimpleBench) | 74.5% ↗ | 81.8% ↗ | -7.3 pt | Muse Spark 1.3 |
| Terminal-Bench 2.1 | 82.9% ↗ | 88.8% ↗ | -5.9 pt | Muse Spark 1.3 |
| Terminal-Bench 4.0 (Vals AI) | 6.1% ↗ | 24.7% ↗ | -18.6 pt | Muse Spark 1.3 |
| Only Muse Spark 1.2 reports · 7 | ||||
| DeepSWE v1.1 (Datacurve) | 54.9% ↗ | not reported | — | — |
| FrontierSWE V2 (Proximal Labs) | 12% ↗ | not reported | — | — |
| MCP Atlas | 90.3% ↗ | not reported | — | — |
| SAGE (Vals AI) | 47.7% ↗ | not reported | — | — |
| SimpleQA Verified | 60.3% ↗epoch run | not reported | — | — |
| Toolathlon-Verified (HKUST) | 75.9% ↗ | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 60.3% ↗ | not reported | — | — |
| Only Muse Spark 1.3 reports · 13 | ||||
| AutomationBench | not reported | 49.6% ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 38% ↗epoch run | — | — |
| DeepSearchQA | not reported | 90.3% ↗ | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 46.3% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 74.4% ↗epoch run | — | — |
| GDPval-AA v2 | not reported | 1754 ↗ | — | — |
| JobBench | not reported | 64.9% ↗ | — | — |
| MRCR 256K-512K | not reported | 98.5% ↗ | — | — |
| MRCR 512K-1M | not reported | 98.1% ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 25% ↗epoch run | — | — |
| OSWorld 2.0 (partial) | not reported | 66.9% ↗ | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 99.2% ↗epoch run | — | — |
| SWE-Atlas Codebase QnA | not reported | 59.4% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Muse Spark 1.2 minus Muse Spark 1.3 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Muse Spark 1.2 | Muse Spark 1.3 | Edge |
|---|---|---|---|
| Context window | 1.0M tokens | 1.0M tokens | Tie |
| Max output | — | — | — |
| Input price / 1M | $1.25 ↗ | $1.25 ↗ | Tie |
| Output price / 1M | $4.25 ↗ | $4.25 ↗ | Tie |
| Cached input / 1M | $0.15 ↗ | $0.15 ↗ | Tie |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $5.5 | $5.5 | Tie |
| Modalities | text · vision · video · audio | text · vision · video | Muse Spark 1.2 |
| Released | 2026-08-05 | 2026-09-02 | — |
| Cited benchmark scores | 23 | 29 | — |
Reliability
Meta status
Live from /statusMore matchups
Muse Spark 1.2 vs …
Models sharing the most benchmarksMore matchups
Muse Spark 1.3 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Muse Spark 1.2 and Muse Spark 1.3 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.