Claude Fable 5 vs GPT-6.2 Astra
GPT-6.2 Astra leads Claude Fable 5 on coding, 87.0 to 75.9 on the Vibe Coding Index, and GPT-6.2 Astra costs 1.3x less per output token ($40.00 vs $50.00 per million). Updated November 22, 2026; 53 benchmarks have a result for both.
Claude Fable 5 or GPT-6.2 Astra: which to pick
Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 75.9 for Claude Fable 5, though on 4 benchmarks against 5.
Pick Claude Fable 5 for work measured by Vibe Code Bench (Vals AI) or MirrorCode, where it leads.
- Shared results
- 53
- Claude Fable 5 only
- 79
- GPT-6.2 Astra only
- 23
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
GPT-6.2 Astra
GPT-6.2 Astra scores 87.0 against 75.9.
5 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
No call
Only GPT-6.2 Astra has a AA Intelligence Index result, 62.1.
Directional only
Reasoning
GPQA Diamond
No call
Only Claude Fable 5 has a GPQA Diamond result, 93.2%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Fable 5 has a MMLU Pro result, 91.5%.
Directional only
Web design
Design Arena: websites
GPT-6.2 Astra
GPT-6.2 Astra leads, 1333 against 1283.
Same board, both measured
General
Vals Index
No call
Only GPT-6.2 Astra has a Vals Index result, 67.7%.
Directional only
Long documents
Listed context window
GPT-6.2 Astra
GPT-6.2 Astra takes 1.05M tokens against 1M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6.2 Astra
Claude Fable 5 $0.035, GPT-6.2 Astra $0.028 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Claude Fable 5 has the number, $2.16.
Directional only
At a glance
- Coding index
- Claude Fable 5 75.9 from 5 benchmarks, GPT-6.2 Astra 87.0 from 4 benchmarks.
- Widest category
- Reasoning, 19.7 points apart on average over 9 shared benchmarks.
- Widest single gap
- Terminal-Bench-Science (Vals AI): Claude Fable 5 15.7% against GPT-6.2 Astra 71.4%, 55.7 points apart on a 0 to 100 scale.
- Output price
- Claude Fable 5 $50.00 against GPT-6.2 Astra $40.00 per million output tokens, 1.3x apart.
- Context window
- Claude Fable 5 takes 1M tokens, GPT-6.2 Astra 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Claude Fable 5 | Bars | GPT-6.2 Astra | Ties |
|---|---|---|---|---|
| Coding | 4 | 6 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 4 | 5 | ||
| Reasoning | 3 | 5 | 1 tie | |
| Math | 0 | 3 | 1 tie | |
| Knowledge | 3 | 1 | ||
| Web design | 0 | 5 | ||
| General | 10 | 1 | ||
| All shared boards | 24 | 27 | 2 ties |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index46
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 91.9% | Only GPT-6.2 Astra measured |
| LMArena WebDev | 1626 | 1766 | GPT-6.2 Astra |
| LiveCodeBench (Vals AI) | 89.8% | Not measured | Only Claude Fable 5 measured |
| SWE-bench Verified (Vals AI) | 95.0% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench 4 (Vals AI) | 41.4% | 63.1% | GPT-6.2 Astra |
| Vibe Code Bench (Vals AI) | 90.3% | 89.9% | Claude Fable 5 |
| Code migration (Vals AI) | 55.1% | Not measured | Only Claude Fable 5 measured |
| ProgramBench (Vals AI) | 2.0% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench 2.1 (Vals AI) | 80.5% | Not measured | Only Claude Fable 5 measured |
| SciCode | 61.0% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 62.9% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 42.4% | 64.1% | GPT-6.2 Astra |
| FrontierCode 2.0 | Not measured | 49.0% | Only GPT-6.2 Astra measured |
| FrontierCode | 53.5% | Not measured | Only Claude Fable 5 measured |
| FrontierSWE | 47.0% | Not measured | Only Claude Fable 5 measured |
| GSO-Bench | 78.4% | Not measured | Only Claude Fable 5 measured |
| WeirdML | 91.9% | Not measured | Only Claude Fable 5 measured |
| MirrorCode | 70.0% | 67.8% | Claude Fable 5 |
| ALE-Bench | 2041.3 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: coding | 1519 | 1450 | Claude Fable 5 |
| LMArena Text: software and IT | 1512 | 1443 | Claude Fable 5 |
| LMArena WebDev: React | 1633 | 1770 | GPT-6.2 Astra |
| LMArena WebDev: HTML | 1651 | 1854 | GPT-6.2 Astra |
| LMArena WebDev: from an image | 1620 | Not measured | Only Claude Fable 5 measured |
| SWE Atlas: Codebase QnA | 39.0% | Not measured | Only Claude Fable 5 measured |
| SWE Atlas: Refactoring | 54.8% | Not measured | Only Claude Fable 5 measured |
| SWE Atlas: Test Writing | 55.6% | Not measured | Only Claude Fable 5 measured |
| GameBench 2 | 75.1 | 97.2 | GPT-6.2 Astra |
| Terminal-Bench 2.1 | 83.8% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench 3.0 | 34.1% | Not measured | Only Claude Fable 5 measured |
| Terminal-Bench 4.0 | 44.5% | Not measured | Only Claude Fable 5 measured |
| Neither model has a result on the other 21 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 62.1 | Only GPT-6.2 Astra measured |
| Epoch Capabilities Index | 164.2 | 172.8 | GPT-6.2 Astra |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index45
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1230 | 1334 | GPT-6.2 Astra |
| Design Arena: agents building full-stack apps | 1258 | 1351 | GPT-6.2 Astra |
| Design Arena: agents building mobile apps | 1255 | 1305 | GPT-6.2 Astra |
| Design Arena: agents building Android apps | 1275 | 1257 | Claude Fable 5 |
| Design Arena: agents building Godot games | 1261 | Not measured | Only Claude Fable 5 measured |
| Design Arena: agents building games | 1231 | Not measured | Only Claude Fable 5 measured |
| Design Arena: HTML slides | 1178 | Not measured | Only Claude Fable 5 measured |
| Design Arena: PowerPoint slides | 1269 | Not measured | Only Claude Fable 5 measured |
| Design Arena: spreadsheets | Not measured | 1278 | Only GPT-6.2 Astra measured |
| LiveBench research | Not measured | 91.5% | Only GPT-6.2 Astra measured |
| LiveBench browser | Not measured | 93.0% | Only GPT-6.2 Astra measured |
| LiveBench long agentic | Not measured | 70.8% | Only GPT-6.2 Astra measured |
| Finance Agent (Vals AI) | 56.3% | Not measured | Only Claude Fable 5 measured |
| Tax Agent Bench (Vals AI) | 38.3% | 21.8% | Claude Fable 5 |
| Legal Research Bench (Vals AI) | 49.5% | 42.8% | Claude Fable 5 |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 16.3% | Only GPT-6.2 Astra measured |
| Excel Modeling Benchmark (Vals AI) | 73.7% | Not measured | Only Claude Fable 5 measured |
| BioMysteryBench (Vals AI) | Not measured | 80.7% | Only GPT-6.2 Astra measured |
| Terminal-Bench-Science (Vals AI) | 15.7% | 71.4% | GPT-6.2 Astra |
| Vals Science Index | Not measured | 68.4% | Only GPT-6.2 Astra measured |
| Robo-bench (Vals AI) | Not measured | 32.7% | Only GPT-6.2 Astra measured |
| τ²-Bench Telecom | 98.5% | Not measured | Only Claude Fable 5 measured |
| APEX-Agents | 63.6% | 72.2% | GPT-6.2 Astra |
| Remote Labor Index | 16.1% | Not measured | Only Claude Fable 5 measured |
| PostTrainBench | 37.3% | Not measured | Only Claude Fable 5 measured |
| SpireClimb | Not measured | 41.3% | Only GPT-6.2 Astra measured |
| Vending-Bench 2 | $5,680 | Not measured | Only Claude Fable 5 measured |
| LMArena Search | 1228 | 1191 | Claude Fable 5 |
| MCP Atlas | 83.3% | Not measured | Only Claude Fable 5 measured |
| τ³-Banking | 39.7% | Not measured | Only Claude Fable 5 measured |
| Neither model has a result on the other 14 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index35
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| GPQA Diamond | 93.2% | Not measured | Only Claude Fable 5 measured |
| LiveBench reasoning | Not measured | 91.6% | Only GPT-6.2 Astra measured |
| MysteryMechanism (Vals AI) | Not measured | 53.1% | Only GPT-6.2 Astra measured |
| MMMU Pro (Vals AI) | 89.3% | Not measured | Only Claude Fable 5 measured |
| GPQA Diamond (Artificial Analysis run) | 92.6% | Not measured | Only Claude Fable 5 measured |
| Humanity's Last Exam (Artificial Analysis run) | 55.5% | Not measured | Only Claude Fable 5 measured |
| HLE-Diamond (Artificial Analysis run) | Not measured | 72.2% | Only GPT-6.2 Astra measured |
| CritPt v2, hard subset (Artificial Analysis run) | Not measured | 71.5% | Only GPT-6.2 Astra measured |
| GPQA Diamond (Epoch AI run) | 85.9% | 96.8% | GPT-6.2 Astra |
| ARC-AGI | 98.5% | 98.5% | Tie |
| ARC-AGI-2 | 89.2% | 95.8% | GPT-6.2 Astra |
| SimpleBench | 81.9% | 77.3% | Claude Fable 5 |
| Chess puzzles | 47.0% | 78.0% | GPT-6.2 Astra |
| Mystery game puzzles | 52.0% | 95.0% | GPT-6.2 Astra |
| EBR-bench | 39.5% | 92.4% | GPT-6.2 Astra |
| LMArena Text: hard prompts | 1508 | 1425 | Claude Fable 5 |
| LMArena Text: expert | 1535 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: hard prompts in English | 1510 | 1426 | Claude Fable 5 |
| LMArena Vision: diagrams | 1354 | Not measured | Only Claude Fable 5 measured |
| LMArena Vision: homework | 1361 | Not measured | Only Claude Fable 5 measured |
| Kagi LLM Benchmark | 91.4% | Not measured | Only Claude Fable 5 measured |
| Humanity's Last Exam (Scale run) | Not measured | 66.9% | Only GPT-6.2 Astra measured |
| GoBench | Not measured | 2493 | Only GPT-6.2 Astra measured |
| Neither model has a result on the other 3 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 100.0% | Tie |
| FrontierMath, tiers 1 to 3 | 87.0% | 92.6% | GPT-6.2 Astra |
| FrontierMath, tier 4 | 90.2% | 100.0% | GPT-6.2 Astra |
| ProofBench | 95.0% | Not measured | Only Claude Fable 5 measured |
| LMArena Text: math | 1516 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: mathematical work | 1516 | Not measured | Only Claude Fable 5 measured |
| MathArena | 46.4% | 92.9% | GPT-6.2 Astra |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index31
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| MMLU Pro | 91.5% | Not measured | Only Claude Fable 5 measured |
| LegalBench (Vals AI) | 88.6% | Not measured | Only Claude Fable 5 measured |
| CorpFin (Vals AI) | 71.8% | Not measured | Only Claude Fable 5 measured |
| TaxEval (Vals AI) | 76.9% | Not measured | Only Claude Fable 5 measured |
| MortgageTax (Vals AI) | 68.9% | Not measured | Only Claude Fable 5 measured |
| MedCode (Vals AI) | 56.1% | Not measured | Only Claude Fable 5 measured |
| AA-Omniscience | Not measured | 50.0 | Only GPT-6.2 Astra measured |
| SimpleQA Verified | 70.7% | 78.5% | GPT-6.2 Astra |
| GeoBench | 89.0% | Not measured | Only Claude Fable 5 measured |
| LMArena Text: science | 1506 | 1415 | Claude Fable 5 |
| LMArena Text: medicine and healthcare | 1480 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: legal and government | 1499 | 1415 | Claude Fable 5 |
| LMArena Text: business and finance | 1479 | 1413 | Claude Fable 5 |
| PRBench Finance | 53.9% | Not measured | Only Claude Fable 5 measured |
| PRBench Legal | 53.9% | Not measured | Only Claude Fable 5 measured |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| Design Arena: websites | 1283 | 1333 | GPT-6.2 Astra |
| Design Arena: UI components | 1303 | 1341 | GPT-6.2 Astra |
| Design Arena: game dev | 1308 | 1501 | GPT-6.2 Astra |
| Design Arena: data viz | 1301 | Not measured | Only Claude Fable 5 measured |
| Design Arena: 3D | 1284 | 1430 | GPT-6.2 Astra |
| Design Arena: all code categories | 1293 | 1389 | GPT-6.2 Astra |
| Design Arena: SVG | 1287 | Not measured | Only Claude Fable 5 measured |
| Design Arena: ASCII art | 1312 | Not measured | Only Claude Fable 5 measured |
| LiveBench design | Not measured | 82.8% | Only GPT-6.2 Astra measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index101
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| Vals Index | Not measured | 67.7% | Only GPT-6.2 Astra measured |
| LiveBench data analysis | Not measured | 84.1% | Only GPT-6.2 Astra measured |
| Public Benefits Bench (Vals AI) | 70.4% | Not measured | Only Claude Fable 5 measured |
| MedScribe (Vals AI) | 88.5% | Not measured | Only Claude Fable 5 measured |
| Vals Multimodal Index | 74.2% | Not measured | Only Claude Fable 5 measured |
| IFBench | 63.5% | Not measured | Only Claude Fable 5 measured |
| Long context reasoning (AA-LCR) | 82.3% | Not measured | Only Claude Fable 5 measured |
| GDPval-AA | 1618 | Not measured | Only Claude Fable 5 measured |
| DTBench | 98.4% | Not measured | Only Claude Fable 5 measured |
| LMCA | 61.2% | Not measured | Only Claude Fable 5 measured |
| Surface Evolver Bench | 95.0% | Not measured | Only Claude Fable 5 measured |
| Furniture assembly | 35.8% | 89.2% | GPT-6.2 Astra |
| LMArena Text | 1492 | 1414 | Claude Fable 5 |
| LMArena Text: instruction following | 1503 | 1423 | Claude Fable 5 |
| LMArena Text: creative writing | 1494 | 1417 | Claude Fable 5 |
| LMArena Vision | 1325 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: English | 1499 | 1417 | Claude Fable 5 |
| LMArena Text: non-English | 1482 | 1404 | Claude Fable 5 |
| LMArena Text: Chinese | 1544 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: German | 1482 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: Japanese | 1505 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: Korean | 1482 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: French | 1510 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: Spanish | 1488 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: Russian | 1501 | 1437 | Claude Fable 5 |
| LMArena Text: Polish | 1488 | Not measured | Only Claude Fable 5 measured |
| LMArena Text: multi-turn | 1504 | 1425 | Claude Fable 5 |
| LMArena Text: longer queries | 1510 | 1431 | Claude Fable 5 |
| LMArena Text: writing and language | 1502 | 1440 | Claude Fable 5 |
| LMArena Text: entertainment and media | 1484 | 1413 | Claude Fable 5 |
| LMArena Vision: English | 1325 | Not measured | Only Claude Fable 5 measured |
| LMArena Vision: Chinese | 1377 | Not measured | Only Claude Fable 5 measured |
| LMArena Vision: reading text | 1336 | Not measured | Only Claude Fable 5 measured |
| LMArena Vision: humor | 1348 | Not measured | Only Claude Fable 5 measured |
| LMArena Vision: creative writing | 1342 | Not measured | Only Claude Fable 5 measured |
| LMArena Document | 1497 | Not measured | Only Claude Fable 5 measured |
| MultiChallenge | Not measured | 73.2% | Only GPT-6.2 Astra measured |
| EQ-Bench 4 | 1313 | Not measured | Only Claude Fable 5 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Fable 5 | GPT-6.2 Astra | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-6.2 Astra costs less
Repository review
50K input + 3K output tokens
GPT-6.2 Astra costs less
Agent loop
200K input + 10K output tokens
GPT-6.2 Astra costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Anthropic Claude Fable 5 | OpenAI GPT-6.2 Astra |
|---|---|---|
| Maker | Anthropic | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | June 9, 2026 | November 17, 2026 |
| Context window | 1M | 1.05M |
| Input $/Mtok | $10.00 | $8.00 |
| Output $/Mtok | $50.00 | $40.00 |
| Coding index | 75.9 | 87.0 |
| Independent measurements behind it | 5 | 4 |
| Cost per solved task | $2.16 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Fable 5 better than GPT-6.2 Astra for coding?
Pick GPT-6.2 Astra for coding: it scores 87.0 on the Vibe Coding Index against 75.9 for Claude Fable 5, though on 4 benchmarks against 5. Pick Claude Fable 5 for work measured by Vibe Code Bench (Vals AI) or MirrorCode, where it leads.
Which is cheaper, Claude Fable 5 or GPT-6.2 Astra?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; GPT-6.2 Astra costs $8.00 and $40.00. Cost per solved task is known for Claude Fable 5 only.
Which has the bigger context window, Claude Fable 5 or GPT-6.2 Astra?
GPT-6.2 Astra, with 1.05M tokens against 1M.
How many benchmarks is the Claude Fable 5 vs GPT-6.2 Astra comparison based on?
53 benchmarks have a result for both models as of November 22, 2026. Claude Fable 5's coding index rests on 5 independent benchmarks and GPT-6.2 Astra's on 4. Where only one model has a result, the page says so instead of naming a winner.