Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.
Specification
most-agreed values
Context
1M
Max output
131K
Released
2026-09-02
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text, image, pdf
Output
text
Price
US dollars per million tokens · most-agreed
Input
$1.25
Output
$4.25
Cache read
$0.15
Quality
6 benchmarks · 2 sources
Composite
84th
percentile of 344 scored models
Rank
55 / 344
±0.10 sd
Evidence
6 × 2
3 or more benchmarks, from 2 or more sources
Effort range
3.9
points between effort settings
reasoning60th2/4 bench
math87th3/7 bench
general94th1/1 bench
By reasoning effort
every setting placed on the same scale as the leaderboard
| Setting | Composite | Benchmarks behind it |
|---|---|---|
| xhigh | 82nd | 6 |
| max | 78th | 4 |
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.2 ±0.6 | xhigh | Epoch AI | 2026-09-17 |
| LiveBench | 82.4 | xhigh · 23 runs | LiveBench | 2026-06-25 |
| FrontierMath-Tiers-1-3-v2 | 74.4 ±2.6 | xhigh | Epoch AI | 2026-09-16 |
| ↳ | 74.0 ±2.6 | max | Epoch AI | 2026-09-18 |
| Settings differ by 0.4 points on this benchmark. | ||||
| Chess Puzzles | 37.0 ±4.9 | max | Epoch AI | 2026-10-13 |
| ↳ | 34.0 ±4.8 | xhigh | Epoch AI | 2026-10-13 |
| Settings differ by 3.0 points on this benchmark. | ||||
| FrontierMath-Tier-4-v2 | 46.3 ±7.9 | max | Epoch AI | 2026-09-18 |
| ↳ | 41.5 ±7.8 | xhigh | Epoch AI | 2026-09-16 |
| Settings differ by 4.8 points on this benchmark. | ||||
| Mystery Game Puzzles | 25.0 ±4.5 | max | Epoch AI | 2026-09-18 |
| ↳ | 14.1 ±3.5 | xhigh | Epoch AI | 2026-09-16 |
| Settings differ by 10.9 points on this benchmark. | ||||
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 17 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| cloudflare-ai-gateway | $1.25 | $4.25 | $0.16 | — | 1M | 131K | — |
| commandcode | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| edenai | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| empiriolabs | $1.25 | $4.25 | $1.00 | — | 1M | 131K | — |
| kilo | $1.25 | $4.25 | $0.15 | — | 1M | 944K | — |
| llmgateway | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| llmgateway-providers | $1.25 | $4.25 | $0.15 | — | 1M | 1M | — |
| meta | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| nano-gpt | $1.25 | $4.25 | $0.15 | — | 1M | 944K | — |
| opencode | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| openrouter | $1.25 | $4.25 | $0.15 | — | 1M | 944K | — |
| opper | $1.25 | $4.25 | $0.15 | — | 1M | 131K | — |
| requesty | $1.25 | $4.25 | $0.15 | — | 1M | 944K | — |
| vercel | $1.25 | $4.25 | $0.15 | — | 1M | 1M | — |
| aimlapi | $1.3225 | $4.4965 | $0.1587 | — | 1M | 131K | — |
| aihubmix | $1.375 | $4.675 | $0.165 | — | 1M | 131K | — |
| kyma · kyma | $1.688 | $5.738 | — | — | 1M | 8K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-03 | openrouter | Cache read | $0.15 |
| 2026-09-19 | openrouter | Cache read | $0.15 |
| 2026-10-10 | openrouter | Cache read | $0.15 |
| 2026-09-03 | openrouter | Input | $1.25 |
| 2026-09-19 | openrouter | Input | $1.25 |
| 2026-10-10 | openrouter | Input | $1.25 |
| 2026-09-03 | openrouter | Output | $4.25 |
| 2026-09-19 | openrouter | Output | $4.25 |
| 2026-10-10 | openrouter | Output | $4.25 |
| 2026-10-13 | aimlapi | Cache read | $0.2063 |
| 2026-10-23 | aimlapi | Cache read | $0.1587 |
| 2026-10-13 | aimlapi | Input | $1.7192 |
| 2026-10-23 | aimlapi | Input | $1.3225 |
| 2026-10-13 | aimlapi | Output | $5.8454 |
| 2026-10-23 | aimlapi | Output | $4.4965 |
More from Meta
most-hosted first