Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 PreviewClaude Opus 5.5 vs Muse Wonder
- Claude Opus 5.5 rates higher on LMArena’s text leaderboard: 1516 against 1501.
- They cost about the same: $4 / $20 and $4 / $20 per million input / output tokens.
- On the 10 Vals AI benchmarks both have taken, Claude Opus 5.5 wins 8 and Muse Wonder wins 2 (Claude Opus 5.5: Vals Index, Terminal-Bench 4.0, Vibe Code Bench).
Side by side
Models (2/6)
- Claude Opus 5.5
- Muse Wonder
| Benchmark | Claude Opus 5.5 | Muse Wonder |
|---|---|---|
| The model | ||
| Lab | Anthropic | Meta |
| Released | 2026-09-22 | 2026-10-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Input, per 1M tokens | $4 | $4 |
| Output, per 1M tokens | $20 | $20 |
| LMArena, people’s votes | ||
| LMArena TextMillions of people compare two anonymous models side by side and vote for the better answer | 1516#2 of 351 | 1501#7 of 351 |
| LMArena WebDevPeople ask two anonymous models to build the same web app, try both and vote for the better one | 1807#1 of 124 | 1651#18 of 124 |
| LMArena VisionSide-by-side votes on prompts that include an image | 1321#2 of 146 | — |
| LMArena CodingLMArena’s text votes on prompts that ask for code | 1545#1 of 346 | 1518#11 of 346 |
| LMArena MathLMArena’s text votes on math prompts | 1518#6 of 339 | 1484#30 of 339 |
| LMArena Creative writingLMArena’s text votes on stories, poems and other creative writing | 1534#1 of 349 | 1493#7 of 349 |
| LMArena Hard promptsLMArena’s text votes on the hardest prompts | 1536#2 of 351 | 1505#12 of 351 |
| LMArena Instruction followingLMArena’s text votes on prompts with specific instructions to follow | 1539#1 of 351 | 1493#10 of 351 |
| Vals AI, overall | ||
| Vals IndexVals AI’s headline score | 70.18%#2 of 27 | 64.88%#6 of 27 |
| Vals AI, coding | ||
| Terminal-Bench 4.0Frontier-difficulty tasks an agent completes in a real terminal | 65.15%#1 of 55 | 49.49%#10 of 55 |
| Vibe Code BenchBuilding complete web apps from a plain-language brief, graded on whether the app works | 90.29%#5 of 115 | 83.09%#20 of 115 |
| Code MigrationRewriting working programs in a new language, graded on whether the new program still passes its tests | 81.01%#2 of 28 | 72.36%#5 of 28 |
| Vals AI, finance | ||
| Finance AgentThe core tasks of a financial analyst, done as an agent with research tools | 67.85%#1 of 31 | 65.55%#6 of 31 |
| Finance IndexVals AI’s composite of finance tasks | 65.59%#1 of 31 | 61.23%#6 of 31 |
| Tax Agent BenchAgentic research into US corporate tax questions | 45.57%#3 of 76 | 36.27%#12 of 76 |
| Vals AI, law | ||
| Legal IndexVals AI’s composite of legal tasks | 72.33%#7 of 28 | 75.26%#3 of 28 |
| Legal Research BenchAgentic US legal research | 50.48%#7 of 85 | 50.48%#6 of 85 |
| Vals AI, medicine | ||
| MedCodeAssigning the right medical billing codes from clinical notes | 49.80%#19 of 107 | — |
| Vals AI, reasoning | ||
| Terminal-Bench ScienceExpert-written scientific research workflows, run by an agent in a terminal | 47.14%#3 of 42 | 22.86%#10 of 42 |
Add models in the compare tool. Vals Index: Claude Opus 5.5 70.18%, Muse Wonder 64.88%.
Questions
Which is better, Claude Opus 5.5 or Muse Wonder?
Claude Opus 5.5 rates higher on LMArena’s text leaderboard: 1516 against 1501. They cost about the same: $4 / $20 and $4 / $20 per million input / output tokens. On the 10 Vals AI benchmarks both have taken, Claude Opus 5.5 wins 8 and Muse Wonder wins 2 (Claude Opus 5.5: Vals Index, Terminal-Bench 4.0, Vibe Code Bench).
Which is cheaper, Claude Opus 5.5 or Muse Wonder?
They cost about the same: $4 / $20 and $4 / $20 per million input / output tokens.