Gemini 3.1 Pro vs Claude Fable 5
Claude Fable 5 was superseded by Claude Fable 5.1. The grades below are the last ones measured, so read them as history rather than as this week's board. See the current roster.
Claude Fable 5 leads overall, C to F. Gemini 3.1 Pro is about 4.4x cheaper per task. By suite, Claude Fable 5 takes real-world issues (C- to D+) and spec planning (C to F). Take Gemini 3.1 Pro anyway when you want the side that is cheaper per task.Measured
Each agent graded on the same private suites, running headless through its own tools. Advantages are oriented so a win is a win, whichever way the metric runs.
Verdict: Claude Fable 5 vs Gemini 3.1 Pro
Claude Fable 5 leads overall, C to F. By suite: Claude Fable 5 takes real-world issues, C- to D+ (60% vs 53% capability); Claude Fable 5 takes spec planning, C to F (63% vs 21% capability); Claude Fable 5 takes vibe coding, C+ to F (74% vs 27% capability). Gemini 3.1 Pro used 15.9% of Google AI Pro's weekly limit per pass, $0.038 per task at the listed price. Claude Fable 5 used 7% of Claude Max 20x's weekly limit per pass, $0.169 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~14% of Claude Max 5x's weekly limit (est, about $0.169 per task). Per task, Gemini 3.1 Pro is about 4.4x cheaper ($0.038 vs $0.169). Claude Fable 5 is faster on real-world issues: a median task takes 2m 37s against 4m 53s. Both finished every task on it. Reasons to pick Gemini 3.1 Pro anyway: it is cheaper per task ($0.038 vs $0.169).
Changed since the previous graded week: Gemini 3.1 Pro: real-world issues usage 9.8% to 14.8% of the weekly limit, spec planning usage 0.9% to 1.1% of the weekly limit, vibe coding usage 1.0% to 0.7% of the weekly limit; Claude Fable 5: real-world issues C+ to C-, vibe coding C to C+.
| Dimension | Gemini 3.1 Pro | Claude Fable 5 | Advantage |
|---|---|---|---|
| Capability | 53% | 60% | +7% (Claude Fable 5 leads) |
| Median task time | 4m 53s | 2m 37s | 2m 16s faster (Claude Fable 5 leads) |
| Approx. task cost | $0.045 / task | $0.153 / task | $0.108 / task lower (Gemini 3.1 Pro leads) |
Gemini 3.1 Pro: 14.8% of Google AI Pro (88.9% / 5h, 14.8% / 1w) per run. Its weekly limit holds 6 of the week's 33.6 5h windows. Claude Fable 5: 33.0% of Claude Max 20x (33.0% / 5h, 5.0% / 1w) per run. Its weekly limit holds about 7 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
| Dimension | Gemini 3.1 Pro | Claude Fable 5 | Advantage |
|---|---|---|---|
| Capability | 21% | 63% | +42% (Claude Fable 5 leads) |
| Median task time | 1m 46s | 5m 32s | 3m 46s faster (Gemini 3.1 Pro leads) |
| Approx. task cost | $0.013 / task | $0.230 / task | $0.217 / task lower (Gemini 3.1 Pro leads) |
Gemini 3.1 Pro: 1.1% of Google AI Pro (6.5% / 5h, 1.1% / 1w) per run. Its weekly limit holds 6 of the week's 33.6 5h windows. Claude Fable 5: 10.0% of Claude Max 20x (10.0% / 5h, 2.0% / 1w) per run. Its weekly limit holds about 7 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
| Dimension | Gemini 3.1 Pro | Claude Fable 5 | Advantage |
|---|---|---|---|
| Capability | 27% | 74% | +47% (Claude Fable 5 leads) |
| Median task time | 4m 38s | 5m 28s | 51s faster (Gemini 3.1 Pro leads) |
| Approx. task cost | $0.034 / task | $0.184 / task | $0.150 / task lower (Gemini 3.1 Pro leads) |
Gemini 3.1 Pro: 0.7% of Google AI Pro (4.4% / 5h, 0.7% / 1w) per good solve. Its weekly limit holds 6 of the week's 33.6 5h windows. Claude Fable 5: 2.8% of Claude Max 20x (2.8% / 5h, 0.4% / 1w) per good solve. Its weekly limit holds about 7 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
Grades and deltas are comparable only within a suite version. See methodology.
Frequently asked questions
Which is better for coding, Gemini 3.1 Pro or Claude Fable 5?
Claude Fable 5 clears higher: tier 4 of 5 versus tier 3 for Gemini 3.1 Pro on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-09-28.
Which is cheaper per task, Gemini 3.1 Pro or Claude Fable 5?
Gemini 3.1 Pro used 15.9% of Google AI Pro's weekly limit per pass, $0.038 per task at the listed price. Claude Fable 5 used 7% of Claude Max 20x's weekly limit per pass, $0.169 per task at the listed price; at the ~$100 tier the board defaults to that is an estimated ~14% of Claude Max 5x's weekly limit (est, about $0.169 per task). Per task, Gemini 3.1 Pro is about 4.4x cheaper ($0.038 vs $0.169).