Muse Spark 1.2 vs Muse Spark 1.3
| Benchmark | Muse Spark 1.2 | Muse Spark 1.3 |
|---|---|---|
| Humanity's Last Exam | 45.5 | 47.5 |
| Arena Text · overall | 1,493 | 1,494 · max |
| Arena Text · Chinese | 1,508.6 | 1,528.3 · max |
| Arena Text · English | 1,484 | 1,493.2 · max |
| Arena Text · French | 1,502.4 | 1,518.8 · max |
| Arena Text · instruction following | 1,461.5 | 1,477.7 · max |
| Arena Text · Korean | 1,473 | 1,497.4 · max |
| Arena Text · non-English | 1,477.9 | 1,481 · max |
| Arena Text · Russian | 1,486.7 | 1,487.9 · max |
| Arena Text · Spanish | 1,488.4 | 1,495.5 · max |
| AA Omniscience Index | 27.2 | 23.1 |
| CritPt (physics) | 17.7 | 26 |
| AA Intelligence Index 4.3.2 | 39.6 | 45.1 |
| AA Omniscience · Accuracy | 45.4 | 41.5 |
| AA Omniscience · Hallucination Rate | 33.3 | 31.5 |
| AA-LCR (long context) 1.1 | 79 | 83 |
| GDP.pdf · AA all-pass | 17.4 | 24.2 |
| MLCR-AA | 31.1 | 43.3 · max |
| Chess Puzzles | 27 | 34 |
| Vals Legal Research Bench | 43.8 | 55.3 · max |
| Arena Text · German | — | 1,523.4 · max |
| Arena Text · Japanese | — | 1,481 · max |
| Arena Text · Polish | — | 1,483.1 · max |
| SimpleQA Verified v1 | 60.3 | — |
| FrontierMath Tier 4 (v2) | — | 41.5 |
| FrontierMath Tiers 1–3 (v2) | — | 74.4 |
| Mystery Game Puzzles | — | 14.1 |
| MathArena · ArXivMath · April 2026 | — | 66.7 |
| MathArena · ArXivMath · August 2026 | — | 45.6 |
| MathArena · ArXivMath · June 2026 | — | 76.7 |
| MathArena · ArXivMath · May 2026 | — | 73.3 |
| MathArena · BrokenArXiv · April 2026 | — | 49.2 |
| MathArena · BrokenArXiv · August 2026 | — | 20.2 |
| MathArena · BrokenArXiv · June 2026 | — | 53.2 |
| MathArena · BrokenArXiv · May 2026 | — | 36.5 |
| MathArena · ArXivLean · June 2026 | — | 28.3 |
| Vals LegalBench | 85.3 | — |
| Vals MedCode | 49.4 | — |
| Vals MedScribe | 90.1 | — |
Specs
| Spec | Muse Spark 1.2 | Muse Spark 1.3 |
|---|---|---|
| Input, per 1M | $1.25 | $1.25 |
| Output, per 1M | $4.25 | $4.25 |
| Cost per task? | $0.97 | $1.37 |
| Speed, tok/s | 355 | 273 |
| Context | 1M | 1M |
| Weights | Closed | Closed |
| Released | Aug 2026 | Sep 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ✓ | ✓ |
| Hear audio | — | — |
| Watch video | ✓ | ✓ |
| Call tools | ✓ | ✓ |
| Force a tool call | — | — |
| Return strict JSON | ✓ | ✓ |
| Search the web | ✓ | ✓ |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Only auto tool_choice is supported›
tool_choice none, required or a named function returns HTTP 400; use auto and handle either a tool call or text.
Search filters return HTTP 400 · Messages API›
allowed_domains, blocked_domains and max_uses return HTTP 400 on Messages search grounding; omit them and use the supported web_search tool.
Manual budget does not set effort · Messages API›
Compatibility budget_tokens does not set reasoning effort; use output_config.effort for depth control.
No Chat Completions search grounding · Chat Completions›
Search grounding is unavailable in Chat Completions; use Responses or Messages.
Reasoning cannot be disabled›
Muse Wonder and Muse Spark do not support disabling reasoning with none; keep reasoning enabled.
Private reasoning still costs tokens›
Private reasoning consumes the output budget and billed output tokens; budget for it even though it is absent from the response body.
Only auto tool_choice is supported›
tool_choice none, required or a named function returns HTTP 400; use auto and handle either a tool call or text.
Reasoning cannot be disabled›
Muse Wonder and Muse Spark do not support disabling reasoning with none; keep reasoning enabled.