Claude Fable 5.1 vs GPT-6.1 Sol
aka Fable 5.1 vs Sol 6.1
- CheaperGPT-6.1 Sol4/4 postsmedian gap ~8.9×
- FasterClaude Fable 5.12/3 postsmedian gap ~1.2×
As stated in each post, not a controlled benchmark.
Posts
4Waves scene scored by two blind AI judges: GPT-6 Astra, GPT-6.1 Sol, Claude Fable 5.1. Posted by @Capocci_AI.
Construction site build, Sol vs Fable 5.1 on xhigh: GPT-6.1 Sol, Claude Fable 5.1. Posted by @FOMO1qse.
35-second casino heist built from code: Claude Opus 5.5, Claude Fable 5.1, GPT-6.1 Sol, GPT-6.1 Terra. Posted by @thehypedotnews.
Same prompt, separate runs
6Not a controlled comparison: prompts, tools and time differ.
About this matchup
Claude Fable 5.1 vs GPT-6.1 Sol in real side-by-side tests: 4 posts from X where both got the same prompt, with cost or time from every post. Judge coding ability from the actual outputs.
Fable 5.1 vs GPT-6.1 Sol: which is better for coding?
Which is better for coding, Fable 5.1 or GPT-6.1 Sol?
None of the 4 posts here names one of the two as the better result, so judge the outputs above for the work you care about.
Which is cheaper, Fable 5.1 or GPT-6.1 Sol?
4 posts state a cost for both: GPT-6.1 Sol was cheaper in all 4 (median gap ~8.9×). These are the costs the posts state for each run, not list prices.
Which is faster, Fable 5.1 or GPT-6.1 Sol?
3 posts state a time for both: Claude Fable 5.1 was faster in 2; GPT-6.1 Sol in 1 (median gap ~1.2×). Times are as stated in each post; tools and settings differ between posts.
Summary
Claude Fable 5.1 and GPT-6.1 Sol appear together in 4 posts on X: crane lifts a concrete slab to the tenth floor (@FOMO1qse), waves scene scored by two blind AI judges (@Capocci_AI), construction site build, Sol vs Fable 5.1 on xhigh (@FOMO1qse) and 35-second casino heist built from code (@thehypedotnews). 2 of them are part of larger lineups with other models. In @thehypedotnews's 35-second casino heist built from code, GPT-6.1 Sol was cheaper ($8.00 vs $75.00), as reported in the post. In @FOMO1qse's construction site build, Sol vs Fable 5.1 on xhigh, GPT-6.1 Sol was cheaper ($0.43 vs $5.68) and Claude Fable 5.1 was faster (15 min vs 18 min), as reported in the post. In @FOMO1qse's crane lifts a concrete slab to the tenth floor, GPT-6.1 Sol was cheaper ($0.65 vs $5.41) and Claude Fable 5.1 was faster (17 min vs 20 min), as reported in the post. In @Capocci_AI's waves scene scored by two blind AI judges, GPT-6.1 Sol was cheaper ($1.61 vs $13.50) and faster (24 min vs 47.4 min), as reported in the post.
What each post reports
| Post | Model | Time | Cost | Tokens | Post |
|---|---|---|---|---|---|
| Crane lifts a concrete slab to the tenth floor@FOMO1qse | GPT-6.1 Sol | 20 min | $0.65 | — | X post |
| Crane lifts a concrete slab to the tenth floor@FOMO1qse | Claude Fable 5.1 | 17 min | $5.41 | — | X post |
| Waves scene scored by two blind AI judges@Capocci_AI | GPT-6.1 Sol | 24 min | $1.61 | — | X post |
| Waves scene scored by two blind AI judges@Capocci_AI | Claude Fable 5.1 | 47.4 min | $13.50 | — | X post |
| Construction site build, Sol vs Fable 5.1 on xhigh@FOMO1qse | GPT-6.1 Sol | 18 min | $0.43 | — | X post |
| Construction site build, Sol vs Fable 5.1 on xhigh@FOMO1qse | Claude Fable 5.1 | 15 min | $5.68 | — | X post |
| 35-second casino heist built from code@thehypedotnews | Claude Fable 5.1 | — | $75.00 | — | X post |
| 35-second casino heist built from code@thehypedotnews | GPT-6.1 Sol | — | $8.00 | — | X post |
Only numbers stated in each post. “—” means the post doesn’t say.
Quick facts
| Model | Vendor | Released | Input $/1M tokens | Output $/1M tokens | Context (tokens) | Official links |
|---|---|---|---|---|---|---|
| Claude Fable 5.1 | Anthropic | |||||
| GPT-6.1 Sol | OpenAI | 2026-09-29 | $2prompts up to 272K tokens | $10prompts up to 272K tokens | 1,050,000 | Model docs |
From the vendors’ official pages, checked 2026-10-19. Standard API list prices; each value links its source and a blank cell could not be verified. The archive’s own numbers above are what each post states.















