Skip to main content

Muse Spark 1.2 vs Muse Spark 1.3

SpecMuse Spark 1.2Muse Spark 1.3
Respan Index45.9 (#53)55.8 (#26)
Blended price per 1M$2$2
Context window1.0M1.0M
Input / output per 1M$1.25 / $4.25$1.25 / $4.25
Max output--
Inputtext, vision, video, audiotext, vision, video
Released2026-08-052026-09-02

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryMuse Spark 1.2Muse Spark 1.3
Coding21.8 Muse Spark 1.240.8 Muse Spark 1.3
Agents58.7 Muse Spark 1.265.6 Muse Spark 1.3
Reasoning47.1 Muse Spark 1.257.4 Muse Spark 1.3
Knowledge & language58.3 Muse Spark 1.261.8 Muse Spark 1.3
Human preference54.7 Muse Spark 1.257.6 Muse Spark 1.3

How the Respan Index works

Benchmarks both report

Muse Spark 1.3 leads on 16 of the 18 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkMuse Spark 1.2Muse Spark 1.3
LMArena Elo14931498
LMArena Vision (LMArena)12931288
LMArena WebDev (LMArena)15341654
LMArena Agent (LMArena)-0.09110.0111
APEX-Agents (Mercor)45.4%58.6%
Chess Puzzles (Epoch AI)27%37%
FrontierSWE V2 (Proximal Labs)12%25.8%
LiveBench Agentic Coding (LiveBench)57.6%64.1%
LiveBench Coding (LiveBench)77.5%81.1%
LiveBench Data Analysis (LiveBench)76.5%79.6%
LiveBench Instruction Following (LiveBench)74.3%78%
LiveBench Language (LiveBench)78.6%82.8%
LiveBench Mathematics (LiveBench)91.2%95.9%
LiveBench Reasoning (LiveBench)90%89.7%
Terminal-Bench 4.0 (Vals AI)6.1%24.7%
SimpleBench (SimpleBench)74.5%81.8%
DeepSWE v1.159.3%75.4%
Terminal-Bench 2.182.9%88.8%

FAQ

Is Muse Spark 1.2 better than Muse Spark 1.3?

Muse Spark 1.3 scores 55.8 on the Respan Index (#26), ahead of Muse Spark 1.2 at 45.9 (#53). Muse Spark 1.3 leads on 16 of the 18 benchmarks both report.

Which is cheaper, Muse Spark 1.2 or Muse Spark 1.3?

Muse Spark 1.2 lists at $1.25 input and $4.25 output per 1M tokens; Muse Spark 1.3 at $1.25 and $4.25. Both cost about the same per token at list price.

Which has the larger context window?

Both accept 1.0M tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.