Compare / head-to-head
Gemini 3.8 Flashvs
Muse Spark 1.3
Muse Spark 1.3 leads 11 of 20 shared benchmarks. Gemini 3.8 Flash is 1.2x cheaper per token. Both offer a 1.0M-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
8 – 11
Muse Spark 1.3 leads · 1 tied
Cheaper per token
Gemini 3.8 Flash
1.2x cheaper, input + output
Larger context
Tie
both 1.0M tokens
Providers
2 providers
Google · Meta
Gemini 3.8 Flash
Google · released 2026-09-02
textvisionaudio
Muse Spark 1.3
Meta · released 2026-09-02
textvisionvideo
Quality
Benchmark matrix
20 shared · 17 only Gemini 3.8 Flash · 9 only Muse Spark 1.3| Benchmark | Gemini 3.8 Flash | Muse Spark 1.3 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 20 | ||||
| LMArena Elo | 1494.8 ↗ | 1494.3 ↗ | +0.5 | Gemini 3.8 Flash |
| APEX-Agents (Mercor) | 64.3% ↗ | 57.8% ↗ | +6.5 pt | Gemini 3.8 Flash |
| Chess Puzzles (Epoch AI run) | 61% ↗epoch run | 38% ↗epoch run | +23 pt | Gemini 3.8 Flash |
| DeepSWE v1.1 | 73.7% ↗ | 75.4% ↗ | -1.7 pt | Muse Spark 1.3 |
| FrontierMath Tier 4 v2 (Epoch AI run) | 22% ↗epoch run | 46.3% ↗epoch run | -24.3 pt | Muse Spark 1.3 |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 68.4% ↗epoch run | 74.4% ↗epoch run | -6 pt | Muse Spark 1.3 |
| LiveBench Agentic Coding (LiveBench) | 54.2% ↗ | 64.1% ↗ | -9.9 pt | Muse Spark 1.3 |
| LiveBench Coding (LiveBench) | 72.5% ↗ | 81.1% ↗ | -8.6 pt | Muse Spark 1.3 |
| LiveBench Data Analysis (LiveBench) | 54% ↗ | 79.6% ↗ | -25.6 pt | Muse Spark 1.3 |
| LiveBench Instruction Following (LiveBench) | 81.4% ↗ | 78% ↗ | +3.4 pt | Gemini 3.8 Flash |
| LiveBench Language (LiveBench) | 87.8% ↗ | 82.8% ↗ | +5 pt | Gemini 3.8 Flash |
| LiveBench Mathematics (LiveBench) | 91.6% ↗ | 95.9% ↗ | -4.3 pt | Muse Spark 1.3 |
| LiveBench Reasoning (LiveBench) | 89.3% ↗ | 89.7% ↗ | -0.4 pt | Muse Spark 1.3 |
| LMArena Agent (LMArena) | 0.0297 ↗ | 0.04 ↗ | 0 | Tie |
| LMArena Vision (LMArena) | 1290.3 ↗ | 1290.2 ↗ | +0.1 | Gemini 3.8 Flash |
| LMArena WebDev (LMArena) | 1582.7 ↗ | 1656.6 ↗ | -73.9 | Muse Spark 1.3 |
| Mystery Game Puzzles (Epoch AI run) | 47% ↗epoch run | 25% ↗epoch run | +22 pt | Gemini 3.8 Flash |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 98.9% ↗epoch run | 99.2% ↗epoch run | -0.3 pt | Muse Spark 1.3 |
| SimpleBench (SimpleBench) | 82.4% ↗ | 81.8% ↗ | +0.6 pt | Gemini 3.8 Flash |
| Terminal-Bench 4.0 (Vals AI) | 19.2% ↗ | 24.7% ↗ | -5.5 pt | Muse Spark 1.3 |
| Only Gemini 3.8 Flash reports · 17 | ||||
| GPQA Diamond | 95.3% ↗ | not reported | — | — |
| CharXiv Reasoning (no tools) | 86.2% ↗ | not reported | — | — |
| DeepSWE v1.1 (Datacurve) | 73.8% ↗ | not reported | — | — |
| FrontierSWE V2 (Proximal Labs) | 19.6% ↗ | not reported | — | — |
| Furniture Assembly (Epoch AI run) | 31.7% ↗epoch run | not reported | — | — |
| GDPVal-AA v2 | 1545 ↗ | not reported | — | — |
| Harvey's Legal Agent Benchmark | 10% ↗ | not reported | — | — |
| HLE-Verified | 54.9% ↗ | not reported | — | — |
| LABBench2 | 86.2% ↗ | not reported | — | — |
| OSWorld-2.0 | 59% ↗ | not reported | — | — |
| SAGE (Vals AI) | 35.1% ↗ | not reported | — | — |
| SimpleQA Verified | 69.7% ↗epoch run | not reported | — | — |
| Terminal-bench 2.1 | 89.4% ↗ | not reported | — | — |
| Terminal-bench 4.0 | 19.1% ↗ | not reported | — | — |
| Vals Finance Agent v2 | 61.4% ↗ | not reported | — | — |
| Vending-Bench 2 (Andon Labs) | 5093.79 ↗ | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 84.8% ↗ | not reported | — | — |
| Only Muse Spark 1.3 reports · 9 | ||||
| AutomationBench | not reported | 49.6% ↗ | — | — |
| DeepSearchQA | not reported | 90.3% ↗ | — | — |
| 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% ↗ | — | — |
| OSWorld 2.0 (partial) | not reported | 66.9% ↗ | — | — |
| SWE-Atlas Codebase QnA | not reported | 59.4% ↗ | — | — |
| Terminal-Bench 2.1 | not reported | 88.8% ↗ | — | — |
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 Gemini 3.8 Flash 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 | Gemini 3.8 Flash | Muse Spark 1.3 | Edge |
|---|---|---|---|
| Context window | 1.0M tokens | 1.0M tokens | Tie |
| Max output | 66K tokens | — | — |
| Input price / 1M | $0.75 ↗ | $1.25 ↗ | Gemini 3.8 Flash |
| Output price / 1M | $3.75 ↗ | $4.25 ↗ | Gemini 3.8 Flash |
| Cached input / 1M | $0.075 ↗ | $0.15 ↗ | Gemini 3.8 Flash |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $4.5 | $5.5 | Gemini 3.8 Flash |
| Modalities | text · vision · audio | text · vision · video | Tie |
| Released | 2026-09-02 | 2026-09-02 | — |
| Cited benchmark scores | 43 | 29 | — |
Reliability
Provider status
Live from /statusMore matchups
Gemini 3.8 Flash 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 Gemini 3.8 Flash and Muse Spark 1.3 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.