BlogLLM Benchmark Scores as Published, Side by Side
Claude Fable 5.2 vs Gemini 4 Carbon: Benchmark Scores Side by Side
Every test both Claude Fable 5.2 by Anthropic and Gemini 4 Carbon by Google have a published number for: 10 tests in 5 areas. Each number is as printed, with the page it comes from and the setup it was run under; nothing is ranked or rescored, and two numbers under different setups are two results, not one comparison.
Models Claude Fable 5.2, Gemini 4 CarbonData version 2026111201, as of 2026-11-12
Model
What each developer prints on its own model page; a dash means it prints none.
| Fact | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| Released | 2026-11-09 | 2026-10-28 |
| Context window | 1M | 1.05M |
| Max output | 128K | 65.5K |
| Input | Text, image | Text, image, video, audio, PDF |
| Output | Text | Text |
| Weights | Closed | Closed |
Knowledge Work
| Test | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| AutomationBench | 51.4%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 53.9%Google, Gemini 4 Carbon, checked 2026-10-29; printed on 1 more page |
| AA-Briefcase v1.1 | 1785Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 1611Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09 |
| AA-Briefcase v1.1 (rubric pass rate) | 66.1%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 61.8%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09 |
Agentic Coding
| Test | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| Terminal-Bench 4.0 | 67.6%Anthropic, System Card: Claude Fable 5.2 & Claude Mythos 5.2, checked 2026-11-09Effort: max; Harness: Claude Code --bare; Fallback: safeguards on; the fallback answered 2.5% of requests, affecting 10% of trials | 65.2%Google, Gemini 4 Carbon, checked 2026-10-29 |
| FrontierSWE v2 | 69.9%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 61.2%Google, Gemini 4 Carbon, checked 2026-10-29; printed on 1 more page |
| Vibe Code Bench | ||
| FrontierCode 2.0 | 49.1%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more pageEffort: xhigh |
Science and Math
| Test | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| Terminal-Bench Science 0.1 | 74.6%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 64.8%Google, Gemini 4 Carbon, checked 2026-10-29; printed on 1 more page |
Computer Use
| Test | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| OSWorld 2.1 offline subset | 89.3%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page |
Multimodal Understanding
| Test | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| Chartography (no tools) | 73.4%Anthropic, Claude Fable 5.2 and Mythos 5.2, checked 2026-11-09; printed on 1 more page | 72.5%Google, Gemini 4 Carbon, checked 2026-10-29; printed on 1 more page |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked; a dash means the developer lists that price in another unit or not at all. Cost per task is measured by Vals AI.
| Price | Claude Fable 5.2 | Gemini 4 Carbon |
|---|---|---|
| Input | $8Anthropic, Anthropic API pricing, checked 2026-11-09 | |
| Cached input | $0.2Anthropic, Anthropic API pricing, checked 2026-11-09 | |
| Cache write | - | |
| Cache write, 1 hour | $16Anthropic, Anthropic API pricing, checked 2026-11-09 | - |
| Output | $40Anthropic, Anthropic API pricing, checked 2026-11-09 | |
| Input, batch | $4Anthropic, Anthropic API pricing, checked 2026-11-09 | |
| Cached input, batch | - | |
| Output, batch | $20Anthropic, Anthropic API pricing, checked 2026-11-09 | |
| Input, flex | - | |
| Cached input, flex | - | |
| Output, flex | - | |
| Input, fast | $16Anthropic, Anthropic API pricing, checked 2026-11-09 | - |
| Output, fast | $80Anthropic, Anthropic API pricing, checked 2026-11-09 | - |
| Input, priority | - | |
| Cached input, priority | - | |
| Output, priority | - | |
| Cost per taskVals AI, USD, lower is better |
Claude Fable 5.2 has 36 more and Gemini 4 Carbon has 15 more: tests only one of them has a number for, listed on its model page.
More side by side
Every number here is read from the same catalog file as the benchmark table, data version 2026111201. A pair has a page when both models are their developers' flagships and both have a number for at least 10 of the same tests. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.