Muse SparkvsClaude Opus 4.6
Across 9 shared benchmarks, Claude Opus 4.6 leads overall: Muse Spark wins 1, Claude Opus 4.6 wins 8, with 0 ties and an average score difference of -23.30.
Muse Spark
Facebook AI研究实验室 · 2026-04-08 · Reasoning model
Claude Opus 4.6
Anthropic · 2026-02-05 · Reasoning model
Muse Spark1 win(11%)(89%)8 winsClaude Opus 4.6
Benchmark scores
Grouped by capability, sorted by largest gap within each. 9 shared benchmarks.
General Knowledge
Claude Opus 4.6 2/3| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 42.5025 / 59Thinking (No Tools) | 66.3015 / 59Extended (no tools) | -23.80 |
| HLE | 584 / 157深度思考（无工具、并行） | 5311 / 157Extended (with tools, internet) | +5 |
| GPQA Diamond | 89.5022 / 178Thinking (No Tools) | 91.3114 / 178Extended (no tools) | -1.81 |
Math and Reasoning
Claude Opus 4.6 2/2| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 14.6023 / 80Normal (No Tools) | 22.9012 / 80最高（无工具） | -8.30 |
| FrontierMath | 399 / 60Thinking (No Tools) | 40.707 / 60最高（无工具） | -1.70 |
Agent Level Benchmark
Claude Opus 4.6 1/1| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 9220 / 35Thinking (With Tools) | 99.252 / 35Extended (with tools) | -7.25 |
AI Agent - Tool Usage
Claude Opus 4.6 1/1| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 5924 / 46Thinking (With Tools) | 65.4011 / 46Extended (with tools) | -6.40 |
Coding and Software Engineer
Claude Opus 4.6 1/1| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| SWE-bench Verified | 77.4024 / 108Thinking (With Tools) | 80.849 / 108Extended (with tools) | -3.44 |
Productivity Knowledge
Claude Opus 4.6 1/1| Benchmark | Muse Spark | Claude Opus 4.6 | Diff |
|---|---|---|---|
| GDPval-AA | 1,4445 / 21Thinking (With Tools) | 1,6063 / 21Extended (with tools, internet) | -162 |
Specs
| Field | Muse Spark | Claude Opus 4.6 |
|---|---|---|
| Publisher | Facebook AI研究实验室 | Anthropic |
| Release date | 2026-04-08 | 2026-02-05 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 262K | 1000K |
| Max output | Not available | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Muse Spark | Claude Opus 4.6 |
|---|---|---|
| Text input | Not public | $0.5 / 1M tokens |
| Text output | Not public | $25 / 1M tokens |
| Cache read | Not public | $0.5 / 1M tokens |
| Cache write | Not public | $10 / 1M tokens |
One or both models have incomplete public pricing.
Summary
- Claude Opus 4.6leads in:General Knowledge (2/3), Math and Reasoning (2/2), Agent Level Benchmark (1/1), AI Agent - Tool Usage (1/1), Coding and Software Engineer (1/1), Productivity Knowledge (1/1)
On average across the 9 shared benchmarks, Claude Opus 4.6 scores 23.30 higher.
Largest single-benchmark gap: GDPval-AA — Muse Spark 1,444 vs Claude Opus 4.6 1,606 (-162).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.