Side-by-side comparison · models
Muse Spark 1.1 vs Muse Spark 1.2
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
Muse Spark 1.1
Open Llama instruction model for multilingual chat, reasoning, and coding.
Models
Muse Spark 1.2
Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks.
Checked facts
Line by line
| Measure | Muse Spark 1.1 | Muse Spark 1.2 |
|---|---|---|
| Input price | $1.25 per million tokens | $1.25 per million tokens |
| Output price | $4.25 per million tokens | $4.25 per million tokens |
| Price on meta | $0.150 per million cached input tokens | $0.150 per million cached input tokens |
| Price on meta | $1.25 per million input tokens | $1.25 per million input tokens |
| Price on meta | $4.25 per million output tokens | $4.25 per million output tokens |
| Latest version | meta/muse-spark-1.1-20260709 | meta/muse-spark-1.2-20260805 |
| Time to first token (ms) | 12,450 | 11,290 |
| Time to first token (ms) on meta | 1,982 | 3,365 |
| Output speed (tokens a second) | 258.5 | 284.1 |
| Output speed (tokens a second) on meta | 203 | 140 |
| Tokens served on OpenRouter in a day | 551.2M (for 10 Nov 2026) | 697.6M (for 10 Nov 2026) |
| Tool calls on OpenRouter in a day | 6.6K (for 10 Nov 2026) | 5.4K (for 10 Nov 2026) |
| Artificial Analysis Intelligence Index | 33.7 | 39.6 |
| Artificial Analysis Coding Index | 71.3 | 72.2 |
| Artificial Analysis Agentic Index | 25.8 | 43.2 |
| LMArena text rating (Elo) | 1,479 | 1,484 |
| LMArena web development rating (Elo) | 1,541 | 1,534 |
| LMArena vision rating (Elo) | 1,292 | 1,305 |
| Design Arena rating (Elo) | 1,281 | 1,326 |
| Simpleqa verified | 57.8% (for 9 July 2026) | 60.3% (for 5 Aug 2026) |
| Apex agents | 31.8% (for 9 July 2026) | 45.4% (for 5 Aug 2026) |
| Chess puzzles | 34% (for 9 July 2026) | 27% (for 5 Aug 2026) |
| Dtbench | 94.4% (for 9 July 2026) | 94.7% (for 5 Aug 2026) |
| Lmca | 49.9% (for 9 July 2026) | 48.4% (for 5 Aug 2026) |
| Proofbench | 39% (for 9 July 2026) | 43% (for 5 Aug 2026) |
| Surface evolver bench | 52.5% (for 9 July 2026) | Not checked yet |
| Frontierswe | Not checked yet | 12% (for 5 Aug 2026) |
| Simplebench | Not checked yet | 74.5% (for 5 Aug 2026) |
| Weirdml | Not checked yet | 60.3% (for 5 Aug 2026) |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.