Claude Fable 5.1 vs GPT-6 Sol
Full profiles: Claude Fable 5.1 (Anthropic) and GPT-6 Sol (OpenAI): every benchmark, price and rank.
Updated · Data from 11 sources
As of 9 Nov 2026, Claude Fable 5.1 is ahead of GPT-6 Sol on quality (66.9 vs 62.9 on the BenchLeader Index), but GPT-6 Sol is 5.0× cheaper at $4.00 vs $20.00 per 1M tokens blended. Claude Fable 5.1 leads in every category we measure both in. GPT-6 Sol streams faster.
Key facts (as of 9 Nov 2026)
- Claude Fable 5.1 scores 66.9 on the BenchLeader Index and GPT-6 Sol scores 62.9 (BenchLeader, 9 Nov 2026).
- Claude Fable 5.1 costs $20.00 per 1M tokens blended (list price, 9 Nov 2026).
- GPT-6 Sol costs $4.00 per 1M tokens blended (list price, 9 Nov 2026).
- GPT-6 Sol streams faster than Claude Fable 5.1 (Artificial Analysis, 7-day median, 9 Nov 2026).
- Of the 55 benchmarks both models have, Claude Fable 5.1 leads 41 (BenchLeader, 9 Nov 2026).
- On Arena WebDev, Claude Fable 5.1 is rated 1744 and GPT-6 Sol 1674 (LMArena, as published by 9 Nov 2026).
Claude Fable 5.1 benchmarks 
GPT-6 Sol benchmarks Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Claude Fable 5.1 | 71 vs 67 category score; Arena WebDev 1744 vs 1674 | Medium |
| Agents & tool use | Either | 60 vs 60 category score; Terminal-Bench 57.3% vs 50.3% | Either |
| Maths | Claude Fable 5.1 | 73 vs 69 category score; FrontierMath Tiers 1–3 90.2% vs 89.8% | Medium |
| Reasoning | Claude Fable 5.1 | 70 vs 66 category score; HLE-Diamond (AA) 60.1% vs 45.2% | Medium |
| Knowledge | Claude Fable 5.1 | 73 vs 62 category score; SimpleQA Verified 70.8% vs 60.7% | High |
| Long documents | Either | 68 vs 67 category score; AA-LCR v1.1 85.3% vs 83.7% | Either |
| Images and other media | Claude Fable 5.1 | 67 vs 61 category score; Arena Vision 1322 vs 1232 | Medium |
| Chat quality (human preference) | Claude Fable 5.1 | 70 vs 64 on Arena votes | Medium |
| Lowest cost | GPT-6 Sol | $4.00 vs $20.00 per 1M blended; agentic coding session $0.160 vs $0.800 | – |
| Fastest first answer | GPT-6 Sol | 72.1 s (GPT-6 Sol (max)) vs 171.7 s (Claude Fable 5.1 (max)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Claude Fable 5.1 (max) leads on quality: 66.9 vs 62.9.
- Claude Fable 5.1 (max) is stronger in agents & tools, coding, human preference, knowledge, long context, maths, multimodal, reasoning.
- GPT-6 Sol (max) is 5.0× cheaper ($4.00 vs $20.00 per 1M blended).
- GPT-6 Sol (max) streams 1.4× faster (121 vs 87 tokens per second).
At a glance
| Spec | Claude Fable 5.1 | GPT-6 Sol |
|---|---|---|
| Developer | Anthropic | OpenAI |
| Released | 1 Sep 2026 | 22 Sep 2026 |
| Context window (tokens) | 1,000,000 | 1,050,000 |
| Compared at | Claude Fable 5.1 (max) | GPT-6 Sol (max) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $10.00 / $50.00 / $0.250 | $2.00 / $10.00 / $0.200 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Claude Fable 5.1 (max) | GPT-6 Sol (max) | Leader |
|---|---|---|---|
| Reasoning | 69.6 | 66.2 | Claude Fable 5.1 |
| Coding | 70.9 | 66.5 | Claude Fable 5.1 |
| Agents & tools | 60.3 | 59.6 | level |
| Maths | 72.5 | 68.5 | Claude Fable 5.1 |
| Knowledge | 72.6 | 61.5 | Claude Fable 5.1 |
| Human preference | 69.9 | 64.2 | Claude Fable 5.1 |
| Multimodal | 67.4 | 60.7 | Claude Fable 5.1 |
| Long context | 67.5 | 66.6 | level |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Instruction following.
ModelIndex
Where sources disagree
Across the 55 benchmarks both have, Claude Fable 5.1 leads 41, GPT-6 Sol leads 13 and 1 is a tie. Every overall index that rates both puts Claude Fable 5.1 ahead.
| Overall index | Claude Fable 5.1 (max) | GPT-6 Sol (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 66.9 | 62.9 | Claude Fable 5.1 |
| AA Intelligence Index v5.0 | 49.2 | 44.2 | Claude Fable 5.1 |
| Arena Text | 1503 | 1457 | Claude Fable 5.1 |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Claude Fable 5.1 (max) | GPT-6 Sol (max) |
|---|---|---|
| Blended price per 1M tokens | $20.00 | $4.00 |
| Input / output per 1M | $10.00 / $50.00 | $2.00 / $10.00 |
| Cached input per 1M | $0.250 | $0.200 |
| Price per task: chat reply | $0.019 | $0.0038 |
| Price per task: summarize a 30-page report | $0.150 | $0.030 |
| Price per task: code edit | $0.135 | $0.027 |
| Price per task: agentic coding session | $0.800 | $0.160 |
| Price per task: structured extraction | $0.030 | $0.0060 |
| Agentic coding session, input mostly cached | $0.361 | $0.079 |
| Cost per benchmark task (AA Index) | $12.99 | $2.47 |
| Output tokens per task (AA Index) | 118k | 54k |
| Of which reasoning | 71k | 35k |
| Time per task (AA Index) | 23 min | 7 min |
| Output speed | 87 tok/s | 121 tok/s |
| Time to first answer | 171.7 s | 72.1 s |
| Time to a complete answer | 403.1 s | 130.2 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $12.99 on Claude Fable 5.1 (max) and $2.47 on GPT-6 Sol (max), and Claude Fable 5.1 (max) wrote 2.2× as many output tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. Claude Fable 5.1 (max) scores 66.9 and takes 172 s to its first answer token; Claude Fable 5.1 scores 66.0 in 3 s, Claude Fable 5.1 (xhigh) scores 64.6 in 59 s. GPT-6 Sol (max) scores 62.9 and takes 72 s to its first answer token; GPT-6 Sol (xhigh) scores 58.6 in 24 s, GPT-6 Sol (high) scores 58.2 in 9 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
142 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Claude Fable 5.1 (max) | GPT-6 Sol (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 66.9 | 62.9 |
| Agents & tools score | BenchLeader | 60.3 | 59.6 |
| Coding score | BenchLeader | 70.9 | 66.5 |
| Human preference score | BenchLeader | 69.9 | 64.2 |
| Knowledge score | BenchLeader | 72.6 | 61.5 |
| Long context score | BenchLeader | 67.5 | 66.6 |
| Maths score | BenchLeader | 72.5 | 68.5 |
| Multimodal score | BenchLeader | 67.4 | 60.7 |
| Reasoning score | BenchLeader | 69.6 | 66.2 |
| Blended price $/M | BenchLeader | $20.00 | $4.00 |
| Output speed | BenchLeader | 87 tok/s | 121 tok/s |
| Time to first answer | BenchLeader | 171.7 s | 72.1 s |
| Context window | BenchLeader | 1,000,000 tokens | 1,050,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | – | 94.3% |
| Humanity's Last Exam | Scale AI / CAIS | 46.5%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1521 | 1483 |
| LiveBench 2.0 Reasoning | LiveBench | 91.7%† | – |
| LiveBench 1.0 Reasoning | LiveBench | 91.7% | 88.7% |
| GPQA Diamond (AA) | Artificial Analysis | 93.7% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 59.1% | 47.9% |
| HLE-Diamond (AA) | Artificial Analysis | 60.1% | 45.2% |
| GPQA Diamond (Vals) | Vals AI | 93.4%† | – |
| ARC-AGI-1 | ARC Prize | 97.5% | 95.5% |
| ARC-AGI-2 | ARC Prize | 90.0% | 89.6% |
| ARC-AGI-3 | ARC Prize | – | 23.0% |
| CritPt | Artificial Analysis | 29.7% | 30.9% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 52.5% | 66.5% |
| MysteryMechanism | Vals AI | 47.8%† | 30.2% |
| Chess Puzzles | Epoch AI Benchmarking Hub | 48.0% | 56.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 58.0% | 56.0% |
| LMCA | Epoch AI Benchmarking Hub | 65.5%† | 59.1% |
| DTBench | Epoch AI Benchmarking Hub | 97.6%† | 97.3% |
| Coding | |||
| SciCode | SciCode | 63.1% | 57.6% |
| WeirdML | WeirdML | 92.9% | – |
| FrontierCode | Cognition | 50.9%† | 49.3% |
| GSO-Bench | GSO-Bench | 88.2%† | – |
| Arena Coding | Arena (LMArena) | 1534 | 1520 |
| Arena WebDev | Arena (LMArena) | 1744 | 1674 |
Show all 142 benchmarks
| Metric | Publisher | Claude Fable 5.1 (max) | GPT-6 Sol (max) |
|---|---|---|---|
| Coding | |||
| LiveBench 2.0 Coding | LiveBench | 89.7%† | – |
| LiveBench 1.0 Coding | LiveBench | 86.4% | 81.8% |
| LiveCodeBench | Vals AI | 90.5%† | – |
| IOI | Vals AI | 90.8%† | 82.6% |
| Code Migration | Vals AI | 54.6%† | 57.2% |
| ProgramBench | Vals AI | 7.0%† | 2.0% |
| SWE-bench Multimodal | Vals AI | 61.3%† | – |
| Vibe Code Bench 1-100 | Vals AI | 28.0%† | – |
| Vibe Code Bench v1.1 | Vals AI | 90.3%† | 87.8% |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 60.0%† | – |
| SWE Atlas: Refactoring | Scale AI SEAL | 56.7%† | – |
| SWE Atlas: Test Writing | Scale AI SEAL | 67.0%† | – |
| MirrorCode | Epoch AI Benchmarking Hub | 73.3%† | – |
| CursorBench | Cursor | 51.8% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 2143.2† | 2462 |
| FrontierSWE | Epoch AI Benchmarking Hub | 56.3% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 52.0% | 43.9% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 91.4% | – |
| Startrise Frontend | Startrise | 77.1%† | – |
| GameBench 2 Overall | DuelLab GameBench 2 | 87.2† | 72.9† |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 94.1† | 79.3† |
| Design Arena Website | Design Arena | 1306† | 1285† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 49.0% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 31.6% | 35.6% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 57.3% | 50.3% |
| Remote Labor Index | Scale AI / CAIS | 17.9%† | – |
| APEX-Agents | Mercor | 68.6%† | 54.3% |
| Arena Agent | Arena (LMArena) | 14.3 | 7 |
| LiveBench Long Agentic | LiveBench | 74.2%† | – |
| LiveBench Browser | LiveBench | 89.8%† | – |
| LiveBench Agentic Coding | LiveBench | 66.1% | 52.9% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 85.0%† | 83.2% |
| MCP Atlas | Scale AI SEAL | 87.2%† | – |
| HiL-Bench | Scale AI SEAL | 61.5%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 62.7% | – |
| τ³-Banking (AA) | Artificial Analysis | 47.2% | – |
| ITBench SRE (AA) | Artificial Analysis | 49.5% | 49.4% |
| Analyst Agent (AA) | Artificial Analysis | 57.5% | – |
| CUA-bench | Vals AI | 13.2% | – |
| CyberBench | Vals AI | 70.4%† | 78.0% |
| Finance Agent v2 | Vals AI | 58.9%† | 49.0% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 6.7%† | 1.7% |
| Legal Research Bench | Vals AI | 55.3%† | 28.9% |
| Robo-bench | Vals AI | 5.8% | – |
| SkillsBench | Vals AI | 61.5%† | – |
| SREBench | Vals AI | 22.9%† | 30.5% |
| Tax Agent Bench | Vals AI | 50.8%† | 15.0% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 58.1%† | 44.4% |
| Terminal-Bench Science | Vals AI | 40.0%† | 30.0% |
| Time Horizon Index: KSP | Vals AI | 63.3%† | – |
| Arena Search | Arena (LMArena) | 1229† | 1229 |
| Vending-Bench 2 | Andon Labs | 5421.6† | 14427.9† |
| AutomationBench | Artificial Analysis | 59.4% | 61.6% |
| GDP.pdf (AA) | Artificial Analysis | 26.2% | 25.2% |
| Harvey LAB-AA v1.1 | Harvey | 6.4% | 3.6% |
| AA-Briefcase v1.1 | Artificial Analysis | 1669 | 1476 |
| GDPx | GDPx by SemiAnalysis | 8.5% | – |
| Design Arena Agentic | Design Arena | 1332† | 1212† |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 43.3% | 30.0% |
| CWE-Bench (AA) | Artificial Analysis | 50.8% | 64.2% |
| DeepSec-Bench (AA) | Artificial Analysis | 23.5% | 46.3% |
| CyberGym end-to-end (AA) | Artificial Analysis | 0.8% | 0.0% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 90.2% | 89.8% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 87.8% | 90.0% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 100.0% | 100.0% |
| ProofBench | Vals AI | 100.0% | 83.0%† |
| LiveBench Mathematics | LiveBench | 97.0% | 96.4% |
| Arena Maths | Arena (LMArena) | 1525 | 1440 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 70.8% | 60.7% |
| LiveBench 2.0 Data Analysis | LiveBench | 80.3%† | – |
| LiveBench Research | LiveBench | 88.5%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 80.3% | 81.2% |
| AA-Omniscience v1.1 | Artificial Analysis | 46.3 | 21.4 |
| MMLU-Pro | Vals AI | 92.4%† | – |
| LegalBench | Vals AI | 88.5%† | – |
| TaxEval | Vals AI | 76.0%† | – |
| PRBench Finance | Scale AI SEAL | 50.8%† | – |
| PRBench Legal | Scale AI SEAL | 51.6%† | 38.6%† |
| BioMysteryBench | Vals AI | – | 74.8% |
| Excel Modeling Benchmark | Vals AI | 76.7%† | 71.5% |
| Excel Modeling Benchmark v1.1 | Vals AI | 76.7%† | – |
| MedCode | Vals AI | 53.5%† | 47.1% |
| MedScribe | Vals AI | 91.3%† | 82.0% |
| Public Benefits Bench v1.1 | Vals AI | 74.9%† | 56.6% |
| EBR-bench | Epoch AI Benchmarking Hub | 57.1% | 53.3% |
| GDP.pdf | Epoch AI Benchmarking Hub | 27.6% | 26.4% |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 68.4% | 51.1% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 30.0% | 39.1% |
| HealthBench Professional (AA) | OpenAI | 57.8% | 62.3% |
| Instruction following | |||
| LiveBench Language | LiveBench | 89.5% | 85.3% |
| LiveBench Instruction Following | LiveBench | 73.0% | 68.6% |
| Arena Instruction Following | Arena (LMArena) | 1501 | 1448 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1503 | 1457 |
| Arena Creative Writing | Arena (LMArena) | 1490 | 1430 |
| Arena Multi-turn | Arena (LMArena) | 1487 | 1466 |
| Arena Longer Queries | Arena (LMArena) | 1510 | 1466 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1322 | 1232 |
| LiveBench Design | LiveBench | 81.6%† | – |
| MMMU-Pro | Artificial Analysis | – | 83.0% |
| MMMU-Pro (Vals) | Vals AI | 90.6%† | – |
| MortgageTax | Vals AI | 70.8%† | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 41.9%† | 36.9%† |
| Code2Video-AA | Artificial Analysis | 1325 | – |
| AA-3D | Artificial Analysis | 1139 | 1131 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 85.3% | 83.7% |
| Arena Document | Arena (LMArena) | 1498 | – |
| MLCR | Artificial Analysis | 15.6% | 33.9% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 166.9† | 164.1† |
| LiveBench 2.0 | LiveBench | 84.0%† | – |
| LiveBench 1.0 | LiveBench | 83.4% | 79.3% |
| AA Intelligence Index v5.0 | Artificial Analysis | 49.2 | 44.2 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 53.4 | 47.6 |
| AA Progress Index v1 | Artificial Analysis | 340.8 | 334 |
| Vals RSI Index | Vals AI | 36.1% | 28.1% |
| Vals Science Index | Vals AI | – | 45.0% |
| AA Cyber Index | Artificial Analysis | 25.0 | 36.8 |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 13.7%† | 13.5%† |
Data as of 2026-11-09. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- GPT-6 Sol (max): BenchLeader Index 63.5 → 62.9.
- Claude Fable 5.1 (max): 1 score revision.
- GPT-6 Sol (max): 2 score revisions.
- Claude Fable 5.1 (max): 1 score revision.
- GPT-6 Sol (max): 1 score revision.
- Claude Fable 5.1 (max): 1 score revision.
- GPT-6 Sol (max): 1 score revision.
- Claude Fable 5.1 (max): 1 score revision.
- GPT-6 Sol (max): 6 new results (AA Intelligence Index v5.0, AA-Omniscience v1.1, CritPt v2 hard subset (AA)…).
- GPT-6 Sol (max): 1 score revision.
Related comparisons
- Claude Fable 5.1 vs Claude Fable 5.2
- Claude Fable 5.1 vs Claude Opus 5.5
- Claude Fable 5.1 vs Gemini 4 Carbon
- GPT-6 Sol vs Claude Fable 5.2
- GPT-6 Sol vs Claude Opus 5.5
- GPT-6 Sol vs Gemini 4 Carbon
- GPT-6 Sol vs GPT-6 Astra
- Claude Opus 5 vs GPT-6 Sol
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Claude Fable 5.1 vs GPT-6 Sol: questions
- Is Claude Fable 5.1 better than GPT-6 Sol?
- Claude Fable 5.1 (max) leads on quality: 66.9 vs 62.9. The BenchLeader Index combines every independent quality benchmark; Claude Fable 5.1 (max) is ahead overall as of 9 Nov 2026, but check the category scores for your use.
- Claude Fable 5.1 vs GPT-6 Sol: which is cheaper, and will that change?
- GPT-6 Sol is cheaper today: $4.00 against $20.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Claude Fable 5.1 or GPT-6 Sol better for coding?
- Claude Fable 5.1 scores higher in coding (71 vs 67 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1744 for Claude Fable 5.1 against 1674 for GPT-6 Sol.
- Which is faster, Claude Fable 5.1 or GPT-6 Sol?
- GPT-6 Sol streams faster (121 vs 87 output tokens per second, Artificial Analysis 7-day median).
- Is Claude Fable 5.1 or GPT-6 Sol better for agents and tool use?
- They are level in agentic tasks (60 vs 60 on the category index). On Terminal-Bench, the top shared agents & tools benchmark, it is 57.3% for Claude Fable 5.1 against 50.3% for GPT-6 Sol.
More questions (4)
- Which is better for long documents?
- They are level on long-context benchmarks (68 vs 67). Both take about 1.1M tokens: Claude Fable 5.1 1,000,000, GPT-6 Sol 1,050,000.
- Which has lower latency?
- GPT-6 Sol answers first: 72.1 s against 171.7 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Claude Fable 5.1 vs GPT-6 Sol?
- Across the 55 benchmarks both have, Claude Fable 5.1 leads 41, GPT-6 Sol leads 13 and 1 is a tie. Every overall index that rates both puts Claude Fable 5.1 ahead.
- When was GPT-6 Sol released?
- GPT-6 Sol was released on 22 Sep 2026.