Compare / Claude Fable 5.2 vs GPT-6 Astra
Claude Fable 5.2 vs GPT-6 Astra
Which is cheaper, which is more capable, and which does better at each kind of work, from 67 results on 4 sources that measured both models.
Of the 67 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 10 November 2026 (catalogue release 2026-11-10-cad1a1ab581e). See also Anthropic vs OpenAI, every model from both labs.
Claude Fable 5.2 or GPT-6 Astra? The short answer
Pick Claude Fable 5.2 if quality matters most: it scores higher (Spring Prompt overall 90 against 78). Pick GPT-6 Astra if cost matters more: it costs 2.4 times less per product listing on CatalogBench.
- Spring Prompt overall: Claude Fable 5.2 90, GPT-6 Astra 78 (out of 100; how the score works).
- GPT-6 Astra is cheaper to run: $0.037 against $0.088 per product listing on CatalogBench (2.4 times less).
- Claude Fable 5.2 is newer: released 9 Nov 2026, against 3 Sep 2026 for GPT-6 Astra.
- On our own benchmarks, on the results both models have: Claude Fable 5.2 does better for marketing planning; GPT-6 Astra for product listings.
Head to head
Every fact side by side
| Claude Fable 5.2 | GPT-6 Astra | |
|---|---|---|
| Price per million tokens | $8.00 in · $40.00 out | $10.00 in · $50.00 out |
| Blended (3 in : 1 out) | $16.00 | $20.00 |
| Cheapest host, blended | $16.00 (Amazon Bedrock) | $20.00 (Amazon Bedrock) |
| Cost per product listing (CatalogBench) | $0.088 | $0.037 |
| Cached input, per million | $0.20 cached input | $1.00 cached input |
| Intelligence Index | – | 52.7 |
| Spring Prompt overall | 90 | 78 |
| Output speed, tokens a second | – | 72 |
| Context window | 1,000,000 tokens | 1,050,000 tokens |
| Released | 9 Nov 2026 (newer) | 3 Sep 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.
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
- Claude Fable 5.2 review: first on our survey and ad budget tests, and one of the dearest models to run 10 Nov 2026
- Best AI for Product Descriptions: Accuracy, Cost and Real Examples 20 Oct 2026
- GPT-6.1 Terra review: close to the top of our business tests at half the price of Sol 5 Nov 2026
- Gemini 4 Argon review: Opus-level product data at a quarter of the cost, until you ask it to sell 14 Oct 2026
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 | Claude Fable 5.2 | GPT-6 Astra |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 17 · 15 of 17 measured | $0.52 | $0.28 |
| Cost per game · Bullet 60sUS dollars, lower is better 36 · 37 of 37 measured | $0.0800low reasoning | $0.0943 |
| Ladder Elo · Bullet 60sladder Elo, higher is better 15 · 26 of 37 measured | 491 | 323low reasoning |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 28 · 24 of 29 measured | $0.0875 | $0.0371 |
| Reliably publish-ready% of products, higher is better 3 · 1 of 28 measured | 71.4% | 75.0% |
| Reliably publish-ready · sales brief% of products, higher is better 7 · 1 of 28 measured | 44.6% | 69.6% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 29 · 28 of 29 measured | $1.40 | $1.09 |
| 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. 3 · 1 of 29 measured | 1,425 | 1,593 |
| Accurate decks% of tasks, higher is better 5 · 1 of 29 measured | 83.3% | 100.0% |
| ROASBench measured by Spring Prompt · 3 results | ||
| Overall scorescore out of 100, higher is better 1 · 3 of 29 measured | 61.3 | 56.6 |
| Audience scorescore out of 100, higher is better 4 · 3 of 29 measured | 68.5 | 68.6 |
| Business scorescore out of 100, higher is better 1 · 4 of 29 measured | 72.3 | 57.5 |
Show all 67 results (55 more)
| Measure | Claude Fable 5.2 | GPT-6 Astra |
|---|---|---|
| BulletBench measured by Spring Prompt · 7 results | ||
| Games lost on time · Blitz 3+2% of games, lower is better 13 · 9 of 17 measured | 50.0% | 37.5% |
| Games lost on time · Bullet 60s% of games, lower is better 27 · 29 of 37 measured | 62.5% | 75.0%low reasoning |
| Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 17 measured | 0.0% | 0.0% |
| Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 37 measured | 0.0% | 0.0% |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 3 · 5 of 17 measured | 1,100 | 839 |
| Median move time · Blitz 3+2milliseconds, lower is better 9 · 11 of 17 measured | 4.9 s | 5.6 s |
| Median move time · Bullet 60smilliseconds, lower is better 32 · 28 of 37 measured | 3.0 slow reasoning | 2.3 slow reasoning |
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 1 · 1 of 29 measured | 100.0% | 100.0% |
| Channel rules broken% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Channel rules broken · sales brief% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Claims to check% of products, lower is better 1 · 1 of 27 measured | 0.0% | 0.0% |
| Claims to check · sales brief% of products, lower is better 1 · 1 of 27 measured | 0.0% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 29 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 2 · 4 of 29 measured | 98.6% | 98.5% |
| Content quality · sales brief% of checks, higher is better 3 · 1 of 29 measured | 98.6% | 98.9% |
| Decision accuracy% of decisions, higher is better 2 · 4 of 29 measured | 98.7% | 98.5% |
| Failed outputs% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 1 · 12 of 29 measured | 98.1% | 94.2% |
| Invented values% of filled values, lower is better 18 · 19 of 29 measured | 1.9% | 1.9% |
| Missing UK information% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 29 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 5 · 19 of 29 measured | 1.8% | 6.0% |
| Not findable · sales brief% of products, lower is better 5 · 18 of 29 measured | 1.8% | 5.4% |
| Publish-ready listings% of products, higher is better 1 · 4 of 29 measured | 81.0% | 77.4% |
| Publish-ready listings · sales brief% of products, higher is better 8 · 4 of 29 measured | 66.1% | 73.2% |
| Time per productseconds, lower is better 16 · 14 of 29 measured | 17.5 | 12.3 |
| Unsupported claimsclaims per product, lower is better 7 · 3 of 29 measured | 0.23 | 0.02 |
| Unsupported claims% of products, lower is better 6 · 4 of 29 measured | 6.0% | 1.2% |
| Unsupported claims · sales brief% of products, lower is better 10 · 3 of 29 measured | 23.8% | 2.4% |
| Unsupported claims · sales briefclaims per product, lower is better 8 · 1 of 29 measured | 0.46 | 0.05 |
| Wrong attributes% of products, lower is better 1 · 16 of 29 measured | 9.5% | 17.3% |
| Wrong attributes · sales brief% of products, lower is better 3 · 20 of 29 measured | 13.1% | 20.2% |
| Wrong category or variant% of products, lower is better 5 · 5 of 29 measured | 1.8% | 1.8% |
| Wrong category or variant · sales brief% of products, lower is better 5 · 5 of 29 measured | 1.8% | 1.8% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 14 · 1 of 29 measured | 16.7% | 0.0% |
| Clean layout% of tasks, higher is better 3 · 1 of 29 measured | 83.3% | 100.0% |
| Design quality% of the maximum, higher is better 8 · 1 of 29 measured | 77.8% | 88.5% |
| Draft figure quoted% of tasks, lower is better 12 · 1 of 29 measured | 16.7% | 0.0% |
| Findings missing% of tasks, lower is better 20 · 1 of 29 measured | 16.7% | 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. 6 · 1 of 29 measured | 78.0% | 89.7% |
| Layout defects% of tasks, lower is better 3 · 1 of 29 measured | 16.7% | 0.0% |
| Misleading metric used% of tasks, lower is better 26 · 1 of 29 measured | 16.7% | 0.0% |
| Presentable decks% of tasks, higher is better 1 · 2 of 29 measured | 83.3% | 66.7% |
| Recommendation late or wrong% of tasks, lower is better 12 · 1 of 29 measured | 16.7% | 0.0% |
| Slides needing work% of tasks, lower is better 3 · 9 of 29 measured | 16.7% | 33.3% |
| Time per deckseconds, lower is better 23 · 24 of 29 measured | 285 | 311 |
| Unsupported claims% of tasks, lower is better 15 · 1 of 29 measured | 16.7% | 0.0% |
| Unsupported numbers% of tasks, lower is better 24 · 1 of 29 measured | 16.7% | 0.0% |
| ROASBench measured by Spring Prompt · 6 results | ||
| Consistency scorescore out of 100, higher is better 11 · 2 of 29 measured | 31.6 | 44.7 |
| Contribution profitsimulated US dollars, higher is better 1 · 14 of 29 measured | $1,032,894 | $599,139 |
| Cost of a runUS dollars, lower is better 28 · 26 of 29 measured | $1.76 | $1.23 |
| Months over budgetmonths of 12, lower is better 1 · 1 of 29 measured | 0 | 0 |
| Planning scorescore out of 100, higher is better 26 · 8 of 29 measured | 54.2 | 54.9 |
| Return on ad spendprofit per $1 spent, higher is better 1 · 6 of 29 measured | $0.96 | $0.67 |
Compare them with other models
- Anthropic vs OpenAIEvery model from both labs, best against best
- Claude Fable 5.2 vs Claude Opus 5.5Just below overall
- GPT-6 Astra vs Claude Fable 5.1Most intelligent at $10 or more per million tokens
- GPT-6 Astra vs Claude Sonnet 5.5Cheapest with similar intelligence
- GPT-6 Astra vs Muse WonderFastest with similar intelligence
- GPT-6 Astra vs Claude Opus 5.5Most intelligent available today
Quick answers
Is Claude Fable 5.2 better than GPT-6 Astra?
Pick Claude Fable 5.2 if quality matters most: it scores higher (Spring Prompt overall 90 against 78). Pick GPT-6 Astra if cost matters more: it costs 2.4 times less per product listing on CatalogBench.
Which is cheaper, Claude Fable 5.2 or GPT-6 Astra?
GPT-6 Astra is cheaper to run: $0.037 against $0.088 per product listing on CatalogBench (2.4 times less).
Which is better for product listings, Claude Fable 5.2 or GPT-6 Astra?
GPT-6 Astra ranks 1 of 28 models on sale for turning a sparse product feed and photos into listings that can go live, against 5 for Claude Fable 5.2.
Which is better for decks from an analysis, Claude Fable 5.2 or GPT-6 Astra?
GPT-6 Astra ranks 1 of 28 models on sale for turning a finished analysis into a deck you could present as it is, against 3 for Claude Fable 5.2.
Which is better for marketing planning, Claude Fable 5.2 or GPT-6 Astra?
Claude Fable 5.2 ranks 1 of 28 models on sale for planning a year of ad spend without overspending, against 3 for GPT-6 Astra.
Which is better for speed under pressure, Claude Fable 5.2 or GPT-6 Astra?
Claude Fable 5.2 ranks 7 of 31 models on sale for good decisions against a real clock (fast chess), against 18 for GPT-6 Astra.
Which has the bigger context window, Claude Fable 5.2 or GPT-6 Astra?
GPT-6 Astra: 1,050,000 tokens, against 1,000,000 for Claude Fable 5.2.
Run Claude Fable 5.2 and GPT-6 Astra 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.