Compare / Muse Spark 1.3 vs GPT-6.1 Sol

Muse Spark 1.3 vs GPT-6.1 Sol

Which is cheaper, which is more capable, and which does better at each kind of work, from 72 results on 6 sources that measured both models.

MetaMuse Spark 1.3 Released 2 Sep 2026 · $1.25 in · $4.25 out
OpenAIGPT-6.1 Sol Released 29 Sep 2026 · $2.00 in · $10.00 out
Muse Spark 1.3 wins 823 too close to callGPT-6.1 Sol wins 41

Of the 72 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee). See also Meta vs OpenAI, every model from both labs.

Muse Spark 1.3 or GPT-6.1 Sol? The short answer

Pick GPT-6.1 Sol: it scores higher (Spring Prompt overall 80 against 59) and costs less. Pick Muse Spark 1.3 only if speed matters more: it writes faster.

  • Spring Prompt overall: Muse Spark 1.3 59, GPT-6.1 Sol 80 (out of 100; how the score works).
  • GPT-6.1 Sol is cheaper to run: $0.008 against $0.016 per product listing on CatalogBench (49% less).
  • GPT-6.1 Sol scores higher on the Artificial Analysis Intelligence Index: 51.8 against 48.1.
  • Muse Spark 1.3 writes faster: 254 tokens a second against 53.
  • GPT-6.1 Sol is newer: released 29 Sep 2026, against 2 Sep 2026 for Muse Spark 1.3.
  • On our own benchmarks, on the results both models have: GPT-6.1 Sol does better for product listings, decks from an analysis and speed under pressure.
  • On third-party benchmarks, on the results both models have: Muse Spark 1.3 does better for sticking to the facts; GPT-6.1 Sol for agents and tool use, professional work and reasoning and knowledge.

Head to head

59
Spring Prompt overall
80
48.1
Intelligence Index
51.8
$2.00
Price per million tokens, blendedlower is better
$4.00
254
Output speed, tokens a second
53
1,048,576
Context window, tokens
1,050,000

Every fact side by side

Muse Spark 1.3GPT-6.1 Sol
Price per million tokens$1.25 in · $4.25 out$2.00 in · $10.00 out
Blended (3 in : 1 out)$2.00$4.00
Cheapest host, blended–$4.00 (Amazon Bedrock)
Cost per product listing (CatalogBench)$0.016$0.008
Cached input, per million$0.15 cached input$0.10 cached input
Intelligence Index48.151.8
Spring Prompt overall5980
Output speed, tokens a second25453
Context window1,048,576 tokens1,050,000 tokens
Released2 Sep 202629 Sep 2026 (newer)
WeightsClosedClosed
Developer based inthe USthe 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 17 of 32 2 of 32
Decks from an analysisTurning a finished analysis into a deck you could present as it is 25 of 32 5 of 32
User surveysPlanning a user survey and reading its results without being misled not measured 3 of 23
Marketing planningPlanning a year of ad spend without overspending 25 of 32 not measured
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 13 of 159 6 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 8 of 51 9 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 11 of 187 6 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 7 of 169 16 of 169
Sticking to the factsSummarising without inventing things, and factual answers 11 of 139 18 of 139
Speed under pressureGood decisions against a real clock (fast chess) 33 of 35 3 of 35

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.

First slide of Muse Spark 1.3's deck for Hearthside Coffee
Muse Spark 1.3 · deck rating 819 · needs work
First slide of GPT-6.1 Sol's deck for Hearthside Coffee
GPT-6.1 Sol · deck rating 1,391 · presentable as it is

Every slide, side by side →

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.

MeasureMuse Spark 1.3GPT-6.1 Sol
BulletBench measured by Spring Prompt · 3 results
Cost per game · Bullet 60sUS dollars, lower is better 41 · 50 of 59 measured $0.0142minimal reasoning $0.0210
Ladder Elo · Bullet 60sladder Elo, higher is better 56 · 2 of 59 measured 33minimal reasoning 928Ultrafast (low reasoning)
Games lost on time · Bullet 60s% of games, lower is better 58 · 31 of 59 measured 91.7%minimal reasoning 25.0%Ultrafast (low reasoning)
CatalogBench measured by Spring Prompt · 3 results
Cost per productUS dollars, lower is better 20 · 16 of 33 measured $0.0164 $0.0084
Reliably publish-ready% of products, higher is better 15 · 1 of 32 measured 48.2% 75.0%
Reliably publish-ready · sales brief% of products, higher is better 17 · 2 of 32 measured 10.7% 67.9%
DeckBench measured by Spring Prompt · 3 results
Cost per deckUS dollars, lower is better 13 · 21 of 33 measured $0.0463 $0.15
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. 25 · 5 of 33 measured 819 1,391
Accurate decks% of tasks, higher is better 24 · 1 of 33 measured 16.7% 100.0%
APEX-Agents reported by APEX-Agents · 1 result
Tasks passed% of tasks, higher is better 12 · 10 of 37 measured 57.8% 60.0%max reasoning
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 8 · 18 of 387 measured 1,494max reasoning 1,483max reasoning
Overall · Text factualityArena rating, higher is better 13 · 37 of 163 measured 1,480max reasoning 1,463max reasoning
Business, management and finance · TextArena rating, higher is better 4 · 27 of 380 measured 1,505max reasoning 1,475max reasoning
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 10 · 7 of 249 measured 48.1max reasoning 51.8max reasoning
Humanity's Last Exam% of questions, higher is better 13 · 8 of 250 measured 48.7%max reasoning 52.9%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 19 · 6 of 120 measured 33.3%max reasoning 56.1%max reasoning
Show all 72 results (56 more)
MeasureMuse Spark 1.3GPT-6.1 Sol
BulletBench measured by Spring Prompt · 2 results
Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 59 measured 0.0%minimal reasoning 0.0%
Median move time · Bullet 60smilliseconds, lower is better 57 · 37 of 59 measured 4.2 sminimal reasoning 1.4 sUltrafast (low reasoning)
CatalogBench measured by Spring Prompt · 28 results
Channel compliance% of products, higher is better 1 · 1 of 33 measured 100.0% 100.0%
Channel rules broken% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Channel rules broken · sales brief% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Claims to check% of products, lower is better 1 · 1 of 31 measured 0.0% 0.0%
Claims to check · sales brief% of products, lower is better 1 · 1 of 31 measured 0.0% 0.0%
Conflicts caught% of conflicts, higher is better 1 · 1 of 33 measured 100.0% 100.0%
Content quality% of checks, higher is better 12 · 3 of 33 measured 97.7% 98.6%
Content quality · sales brief% of checks, higher is better 10 · 2 of 33 measured 98.2% 98.8%
Decision accuracy% of decisions, higher is better 10 · 3 of 33 measured 97.6% 98.7%
Failed outputs% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Failed outputs · sales brief% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Field accuracy% of missing fields, higher is better 21 · 8 of 33 measured 92.9% 94.4%
Invented values% of filled values, lower is better 7 · 25 of 33 measured 0.7% 2.2%
Missing UK information% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Missing UK information · sales brief% of products, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Not findable% of products, lower is better 23 · 19 of 33 measured 6.0% 4.8%
Not findable · sales brief% of products, lower is better 22 · 22 of 33 measured 5.4% 5.4%
Publish-ready listings% of products, higher is better 15 · 3 of 33 measured 65.5% 78.0%
Publish-ready listings · sales brief% of products, higher is better 15 · 5 of 33 measured 34.5% 72.0%
Time per productseconds, lower is better 29 · 11 of 33 measured 47.5 10.2
Unsupported claimsclaims per product, lower is better 14 · 1 of 33 measured 0.44 0.00
Unsupported claims% of products, lower is better 14 · 1 of 33 measured 14.9% 0.0%
Unsupported claims · sales briefclaims per product, lower is better 15 · 2 of 33 measured 1.50 0.07
Unsupported claims · sales brief% of products, lower is better 16 · 5 of 33 measured 57.7% 3.6%
Wrong attributes% of products, lower is better 16 · 16 of 33 measured 16.7% 16.7%
Wrong attributes · sales brief% of products, lower is better 5 · 20 of 33 measured 14.3% 19.1%
Wrong category or variant% of products, lower is better 29 · 6 of 33 measured 4.2% 1.8%
Wrong category or variant · sales brief% of products, lower is better 27 · 6 of 33 measured 3.6% 1.8%
DeckBench measured by Spring Prompt · 14 results
Caveat dropped% of tasks, lower is better 15 · 1 of 33 measured 16.7% 0.0%
Clean layout% of tasks, higher is better 29 · 3 of 33 measured 0.0% 83.3%
Design quality% of the maximum, higher is better 25 · 12 of 33 measured 56.1% 71.4%
Draft figure quoted% of tasks, lower is better 13 · 1 of 33 measured 16.7% 0.0%
Findings missing% of tasks, lower is better 1 · 1 of 33 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. 24 · 5 of 33 measured 41.8% 81.0%
Layout defects% of tasks, lower is better 29 · 3 of 33 measured 100.0% 16.7%
Misleading metric used% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Presentable decks% of tasks, higher is better 15 · 3 of 33 measured 0.0% 50.0%
Recommendation late or wrong% of tasks, lower is better 28 · 1 of 33 measured 66.7% 0.0%
Slides needing work% of tasks, lower is better 17 · 17 of 33 measured 50.0% 50.0%
Time per deckseconds, lower is better 12 · 22 of 33 measured 89.4 179
Unsupported claims% of tasks, lower is better 15 · 1 of 33 measured 16.7% 0.0%
Unsupported numbers% of tasks, lower is better 1 · 1 of 33 measured 0.0% 0.0%
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 8 results
Confirmed task success · AgentIPS effect estimate, higher is better 18 · 2 of 48 measured 0.05max reasoning, arena agent 0.15max reasoning, arena agent
Creative writing · TextArena rating, higher is better 22 · 21 of 385 measured 1,459max reasoning 1,460max reasoning
Expert prompts · TextArena rating, higher is better 12 · 4 of 338 measured 1,529max reasoning 1,543max reasoning
Instruction following · TextArena rating, higher is better 13 · 10 of 387 measured 1,486max reasoning 1,490max reasoning
Praise over complaint · AgentIPS effect estimate, higher is better 21 · 6 of 48 measured 0.03max reasoning, arena agent 0.26max reasoning, arena agent
Steerability · AgentIPS effect estimate, higher is better 12 · 3 of 48 measured 0.05max reasoning, arena agent 0.11max reasoning, arena agent
Tool grounding · AgentIPS effect estimate, higher is better 14 · 1 of 48 measured 0.00max reasoning, arena agent 0.00max reasoning, arena agent
Writing, literature and language · TextArena rating, higher is better 19 · 17 of 386 measured 1,470max reasoning 1,472max reasoning
Artificial Analysis reported by Artificial Analysis · 4 results
Long-context reasoning (AA-LCR)% of questions, higher is better 16 · 7 of 241 measured 83.0%max reasoning 84.0%low reasoning
Output speedtokens per second, higher is better 9 · 67 of 78 measured 254xhigh reasoning 53.1max reasoning
SciCode% of problems, higher is better 10 · 26 of 121 measured 59.7%xhigh reasoning 55.8%high reasoning
Time to first answer tokenseconds, lower is better 66 · 35 of 78 measured 29.6xhigh reasoning 1.71low reasoning

Quick answers

Is Muse Spark 1.3 better than GPT-6.1 Sol?

Pick GPT-6.1 Sol: it scores higher (Spring Prompt overall 80 against 59) and costs less. Pick Muse Spark 1.3 only if speed matters more: it writes faster.

Which is cheaper, Muse Spark 1.3 or GPT-6.1 Sol?

GPT-6.1 Sol is cheaper to run: $0.008 against $0.016 per product listing on CatalogBench (49% less).

Which is faster, Muse Spark 1.3 or GPT-6.1 Sol?

Muse Spark 1.3 writes about 254 tokens a second on its usual API, against 53 for GPT-6.1 Sol (Artificial Analysis).

Which is better for product listings, Muse Spark 1.3 or GPT-6.1 Sol?

GPT-6.1 Sol ranks 2 of 32 models on sale for turning a sparse product feed and photos into listings that can go live, against 17 for Muse Spark 1.3.

Which is better for decks from an analysis, Muse Spark 1.3 or GPT-6.1 Sol?

GPT-6.1 Sol ranks 5 of 32 models on sale for turning a finished analysis into a deck you could present as it is, against 25 for Muse Spark 1.3.

Which is better for agents and tool use, Muse Spark 1.3 or GPT-6.1 Sol?

GPT-6.1 Sol ranks 6 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 13 for Muse Spark 1.3.

Which is better for professional work, Muse Spark 1.3 or GPT-6.1 Sol?

Muse Spark 1.3 ranks 8 of 51 models on sale for real tasks from banking, consulting and law, business documents and freelance projects, against 9 for GPT-6.1 Sol.

Which has the bigger context window, Muse Spark 1.3 or GPT-6.1 Sol?

GPT-6.1 Sol: 1,050,000 tokens, against 1,048,576 for Muse Spark 1.3.

Run Muse Spark 1.3 and GPT-6.1 Sol on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.