Compare / Gemini 4 Argon vs Muse Wonder
Gemini 4 Argon vs Muse Wonder
Which is cheaper, which is more capable, and which does better at each kind of work, from 92 results on 6 sources that measured both models.
GoogleGemini 4 Argon
MetaMuse Wonder
Gemini 4 Argon wins 3419 too close to callMuse Wonder wins 39
Of the 92 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 14 October 2026 (catalogue release 2026-10-14-7d7cf32c3f6d). See also Google vs Meta, every model from both labs.
Gemini 4 Argon or Muse Wonder? The short answer
They score about the same, so pick Gemini 4 Argon: it costs 3.3 times less per product listing on CatalogBench.
- Spring Prompt overall: Gemini 4 Argon 75, Muse Wonder 71 (out of 100; how the score works).
- Gemini 4 Argon is cheaper to run: $0.011 against $0.036 per product listing on CatalogBench (3.3 times less).
- They score about the same on the Artificial Analysis Intelligence Index (52.6 and 52.4).
- Gemini 4 Argon is newer: released 13 Oct 2026, against 12 Oct 2026 for Muse Wonder.
- On our own benchmarks, on the results both models have: Gemini 4 Argon does better for marketing planning; Muse Wonder for product listings, decks from an analysis and user surveys.
- On third-party benchmarks, on the results both models have: Gemini 4 Argon does better for agents and tool use and reasoning and knowledge.
Head to head
75
Spring Prompt overall
71
52.6
Intelligence Index
52.4
$4.00
Price per million tokens, blendedlower is better
$8.00
1,048,576
Context window, tokens
1,048,576
Every fact side by side
| Gemini 4 Argon | Muse Wonder | |
|---|---|---|
| Price per million tokens | $2.00 in · $10.00 out | $4.00 in · $20.00 out |
| Blended (3 in : 1 out) | $4.00 | $8.00 |
| Cheapest host, blended | – | – |
| Cost per product listing (CatalogBench) | $0.011 | $0.036 |
| Cached input, per million | – | $0.20 cached input |
| Intelligence Index | 52.6 | 52.4 |
| Spring Prompt overall | 75 | 71 |
| Output speed, tokens a second | – | 117 |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Released | 13 Oct 2026 (newer) | 12 Oct 2026 |
| Weights | Closed | Closed |
| Developer based in | the US | the US |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Product listingsTurning a sparse product feed and photos into listings that can go live
7 of 22
4 of 22
Decks from an analysisTurning a finished analysis into a deck you could present as it is
13 of 21
8 of 21
User surveysPlanning a user survey and reading its results without being misled
5 of 10
2 of 10
Marketing planningPlanning a year of ad spend without overspending
8 of 21
11 of 21
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
3 of 166
8 of 166
Reasoning and knowledgeHard questions across science, maths and general knowledge
3 of 200
7 of 200
WritingWhat people prefer in blind comparisons, and judged writing quality
1 of 189
not measured
Sticking to the factsSummarising without inventing things, and factual answers
1 of 153
not measured
Speed under pressureGood decisions against a real clock (fast chess)
20 of 25
12 of 25
Their slides, side by side
The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.
From our research
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Gemini 4 Argon | Muse Wonder |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 10 · 13 of 15 measured | $0.12 | $0.27 |
| Cost per game · Bullet 60sUS dollars, lower is better 10 · 29 of 30 measured | $0.0036 | $0.0663minimal reasoning |
| Ladder Elo · Bullet 60sladder Elo, higher is better 26 · 19 of 30 measured | 147 | 409minimal reasoning |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 12 · 18 of 22 measured | $0.0109 | $0.0361 |
| Reliably publish-ready% of products, higher is better 5 · 3 of 22 measured | 60.7% | 66.1% |
| Reliably publish-ready · sales brief% of products, higher is better 9 · 4 of 22 measured | 17.9% | 51.8% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 10 · 15 of 21 measured | $0.11 | $0.18 |
| Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 13 · 8 of 21 measured | 998 | 1,297 |
| Accurate decks% of tasks, higher is better 13 · 1 of 21 measured | 33.3% | 100.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 8 · 11 of 21 measured | 47.0 | 38.9 |
| Audience scorescore out of 100, higher is better 8 · 13 of 21 measured | 65.5 | 54.7 |
| Business scorescore out of 100, higher is better 8 · 11 of 21 measured | 43.3 | 34.5 |
| SurveyBench measured by Spring Prompt · 3 results | ||
| Analysis ratingrating, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 5 · 2 of 10 measured | 964 | 1,494 |
| Cost per taskUS dollars, lower is better 6 · 9 of 10 measured | $0.13 | $0.35 |
| Analysis scorepoints of 100, higher is better 5 · 2 of 10 measured | 79.6 | 96.2 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 5 · 6 of 244 measured | 52.6high reasoning | 52.4max reasoning |
| Humanity's Last Exam% of questions, higher is better 3 · 9 of 243 measured | 57.1%high reasoning | 50.1%max reasoning |
| Terminal-Bench 4.0% of tasks, higher is better 4 · 7 of 113 measured | 57.1%high reasoning | 52.0%max reasoning |
Show all 92 results (74 more)
| Measure | Gemini 4 Argon | Muse Wonder |
|---|---|---|
| BulletBench measured by Spring Prompt · 7 results | ||
| Games lost on time · Blitz 3+2% of games, lower is better 1 · 15 of 15 measured | 0.0% | 75.0% |
| Games lost on time · Bullet 60s% of games, lower is better 28 · 23 of 30 measured | 87.5% | 58.3%minimal reasoning |
| Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 15 measured | 0.0% | 0.0% |
| Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 30 measured | 0.0% | 0.0%minimal reasoning |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 1 · 14 of 15 measured | 1,147 | 358 |
| Median move time · Blitz 3+2milliseconds, lower is better 5 · 13 of 15 measured | 1.9 s | 7.1 s |
| Median move time · Bullet 60smilliseconds, lower is better 28 · 25 of 30 measured | 2.9 s | 1.9 sminimal reasoning |
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 1 · 15 of 22 measured | 100.0% | 99.4% |
| Channel rules broken% of products, lower is better 1 · 12 of 22 measured | 0.0% | 0.6% |
| Channel rules broken · sales brief% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Claims to check% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Claims to check · sales brief% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 22 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 4 · 3 of 22 measured | 98.2% | 98.3% |
| Content quality · sales brief% of checks, higher is better 4 · 3 of 22 measured | 98.4% | 98.5% |
| Decision accuracy% of decisions, higher is better 5 · 3 of 22 measured | 97.8% | 98.3% |
| Failed outputs% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 3 · 1 of 22 measured | 94.7% | 96.1% |
| Invented values% of filled values, lower is better 14 · 11 of 22 measured | 1.7% | 1.4% |
| Missing UK information% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 22 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 3 · 5 of 22 measured | 2.4% | 3.6% |
| Not findable · sales brief% of products, lower is better 2 · 9 of 22 measured | 1.8% | 3.6% |
| Publish-ready listings% of products, higher is better 5 · 3 of 22 measured | 70.8% | 75.0% |
| Publish-ready listings · sales brief% of products, higher is better 8 · 5 of 22 measured | 35.1% | 67.3% |
| Time per productseconds, lower is better 3 · 8 of 22 measured | 7.30 | 10.2 |
| Unsupported claimsclaims per product, lower is better 6 · 4 of 22 measured | 0.42 | 0.21 |
| Unsupported claims% of products, lower is better 9 · 4 of 22 measured | 16.1% | 7.1% |
| Unsupported claims · sales briefclaims per product, lower is better 12 · 5 of 22 measured | 1.98 | 0.36 |
| Unsupported claims · sales brief% of products, lower is better 12 · 5 of 22 measured | 63.1% | 17.9% |
| Wrong attributes% of products, lower is better 6 · 2 of 22 measured | 15.5% | 13.1% |
| Wrong attributes · sales brief% of products, lower is better 7 · 4 of 22 measured | 15.5% | 14.3% |
| Wrong category or variant% of products, lower is better 4 · 11 of 22 measured | 1.8% | 2.4% |
| Wrong category or variant · sales brief% of products, lower is better 4 · 11 of 22 measured | 1.8% | 2.4% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Clean layout% of tasks, higher is better 13 · 19 of 21 measured | 33.3% | 16.7% |
| Design quality% of the maximum, higher is better 16 · 9 of 21 measured | 55.2% | 67.1% |
| Draft figure quoted% of tasks, lower is better 17 · 1 of 21 measured | 50.0% | 0.0% |
| Findings missing% of tasks, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 13 · 10 of 21 measured | 57.0% | 66.7% |
| Layout defects% of tasks, lower is better 13 · 19 of 21 measured | 66.7% | 83.3% |
| Misleading metric used% of tasks, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Presentable decks% of tasks, higher is better 12 · 10 of 21 measured | 0.0% | 16.7% |
| Recommendation late or wrong% of tasks, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| Slides needing work% of tasks, lower is better 15 · 15 of 21 measured | 66.7% | 66.7% |
| Time per deckseconds, lower is better 6 · 3 of 21 measured | 67.5 | 28.6 |
| Unsupported claims% of tasks, lower is better 11 · 1 of 21 measured | 16.7% | 0.0% |
| Unsupported numbers% of tasks, lower is better 1 · 1 of 21 measured | 0.0% | 0.0% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 5 · 13 of 21 measured | 34.7 | 25.9 |
| Contribution profitsimulated US dollars, higher is better 5 · 12 of 21 measured | $673,561 | $489,176 |
| Cost of a runUS dollars, lower is better 11 · 16 of 21 measured | $0.24 | $0.57 |
| Months over budgetmonths of 12, lower is better 1 · 1 of 21 measured | 0 | 0 |
| Planning scorescore out of 100, higher is better 1 · 5 of 21 measured | 55.4 | 55.0 |
| Return on ad spendprofit per $1 spent, higher is better 8 · 12 of 21 measured | $0.55 | $0.41 |
| SurveyBench measured by Spring Prompt · 17 results | ||
| Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 5 · 3 of 10 measured | 42.2% | 77.0% |
| Leading or double-barrelled questions% of tasks, lower is better 8 · 5 of 10 measured | 66.7% | 44.4% |
| Mistaken next steps% of tasks, lower is better 5 · 1 of 10 measured | 33.3% | 0.0% |
| Numbers right% of answers, higher is better 7 · 1 of 10 measured | 97.1% | 100.0% |
| Objectives unanswerable% of tasks, lower is better 1 · 6 of 10 measured | 0.0% | 11.1% |
| Precision ignored% of tasks, lower is better 4 · 5 of 10 measured | 11.1% | 22.2% |
| Research scorepoints of 100, higher is better 5 · 3 of 10 measured | 83.3 | 93.6 |
| Sound research% of tasks, higher is better 5 · 3 of 10 measured | 0.0% | 33.3% |
| Stakeholder's view presumed% of tasks, lower is better 4 · 4 of 10 measured | 11.1% | 11.1% |
| Survey plan scorepoints of 100, higher is better 4 · 6 of 10 measured | 91.8 | 87.8 |
| Survey structure% of tasks, lower is better 1 · 9 of 10 measured | 0.0% | 33.3% |
| Time per taskseconds, lower is better 3 · 4 of 10 measured | 74.4 | 83.2 |
| Traps handled% of traps, higher is better 5 · 1 of 10 measured | 84.1% | 100.0% |
| Traps missed% of tasks, lower is better 5 · 1 of 10 measured | 77.8% | 0.0% |
| Unsupported findings% of tasks, lower is better 7 · 2 of 10 measured | 88.9% | 11.1% |
| Wrong numbers% of tasks, lower is better 7 · 1 of 10 measured | 33.3% | 0.0% |
| Wrong recommendation% of tasks, lower is better 3 · 1 of 10 measured | 11.1% | 0.0% |
| Artificial Analysis reported by Artificial Analysis · 2 results | ||
| Long-context reasoning (AA-LCR)% of questions, higher is better 44 · 33 of 234 measured | 79.7%high reasoning | 80.7%max reasoning |
| SciCode% of problems, higher is better 3 · 7 of 114 measured | 61.8%high reasoning | 60.3%max reasoning |
Compare them with other models
- Google vs MetaEvery model from both labs, best against best
- Gemini 4 Argon vs Claude Opus 5.5Most intelligent at $3–10 per million tokens
- Gemini 4 Argon vs Claude Sonnet 5.5Cheapest with similar intelligence
- Gemini 4 Argon vs GPT-6 AstraJust above overall
- Muse Wonder vs Claude Opus 5.5Most intelligent at $3–10 per million tokens
- Muse Wonder vs Claude Sonnet 5.5Cheapest with similar intelligence
- Muse Wonder vs GPT-6 SolJust below overall
Quick answers
Is Gemini 4 Argon better than Muse Wonder?
They score about the same, so pick Gemini 4 Argon: it costs 3.3 times less per product listing on CatalogBench.
Which is cheaper, Gemini 4 Argon or Muse Wonder?
Gemini 4 Argon is cheaper to run: $0.011 against $0.036 per product listing on CatalogBench (3.3 times less).
Which is better for product listings, Gemini 4 Argon or Muse Wonder?
Muse Wonder ranks 4 of 22 models on sale for turning a sparse product feed and photos into listings that can go live, against 7 for Gemini 4 Argon.
Which is better for decks from an analysis, Gemini 4 Argon or Muse Wonder?
Muse Wonder ranks 8 of 21 models on sale for turning a finished analysis into a deck you could present as it is, against 13 for Gemini 4 Argon.
Which is better for user surveys, Gemini 4 Argon or Muse Wonder?
Muse Wonder ranks 2 of 10 models on sale for planning a user survey and reading its results without being misled, against 5 for Gemini 4 Argon.
Which is better for marketing planning, Gemini 4 Argon or Muse Wonder?
Gemini 4 Argon ranks 8 of 21 models on sale for planning a year of ad spend without overspending, against 11 for Muse Wonder.
Which is better for agents and tool use, Gemini 4 Argon or Muse Wonder?
Gemini 4 Argon ranks 3 of 166 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 8 for Muse Wonder.
Run Gemini 4 Argon and Muse Wonder on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.