Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Compare
Side by side: capability scores, runtime measurements and every published price component. A value no source returned stays explicitly unavailable.
Choose two to four models, then compare their evidence.
Selected models 2 of 4
Exact source values, missing states, and pair-derived deltas stay separate.
Evidence key: ObservedUnavailable
| Capability | Claude Fable 5.2 | Kimi K3.1 Pro |
|---|---|---|
| Agentic | Unavailable | 82.8 |
| Coding | 88 | Unavailable |
| Knowledge | 94 | 90 |
| Reasoning | 86.8 | 81.6 |
| Multimodal & Grounded | 89.3 | 95.2 |
| Instruction Following | Unavailable | Unavailable |
| Math | 93.8 | Unavailable |
| Overall | 93.2 | 70.6 |
Time to first token and throughput, exactly as the runtime source published them.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Source rates plus a labelled 75% input / 25% output blended cost.
| Metric | Claude Fable 5.2 | Kimi K3.1 Pro |
|---|---|---|
| Input / 1M | $8.00 | $2.00 |
| Output / 1M | $40.00 | $10.00 |
| Cache read / 1M | $0.20 | $0.20 |
| Cache write / 1M | $10.00 | Unavailable |
| Blended / 1M (75% input / 25% output) | $16.00 | $4.00 |
| Cost per successful task (LiveBench) | Unavailable | Unavailable |
| Time to first token | 228.89 s | 158.74 s |
| Throughput | 111 tok/s | 31 tok/s |
| Context window | 1M tokens | 1.05M tokens |
| Release date | 2026-11-09 | 2026-10-23 |
| Access | Proprietary | Open Weight |
Each row comes from the pair comparison source. A null basis or average stays unavailable rather than being inferred from the capability table.
benchlm:comparisons · bench-align-v5 · fetched 2026-11-21 · overall delta 22.60 · overall winner A
| Capability | A average | B average | Delta | Winner | Pair basis |
|---|---|---|---|---|---|
| Agentic | 92.8 | 72.9 | 19.90 | A | Supported · #2/178 · Supported · #18/178 · like-for-like · 26 benchmarks |
| Coding | 92.3 | 74.9 | 17.40 | A | Supported · #1/178 · Supported · #18/178 · like-for-like · 16 benchmarks |
| Knowledge | 92 | 66.6 | 25.40 | Unavailable | Estimated · #1/157 · Estimated · #25/157 · directional · 6 benchmarks |
| Reasoning | 93.6 | 66.3 | 27.30 | A | Supported · #1/161 · Supported · #27/161 · like-for-like · 3 benchmarks |
| Multimodal & Grounded | 88 | 70.4 | 17.60 | A | Supported · #1/116 · Supported · #15/116 · like-for-like · 19 benchmarks |
| Instruction Following | 87.4 | 64.7 | 22.70 | A | Supported · #1/142 · Supported · #22/142 · like-for-like · 0 benchmarks |
| Math | 86.9 | Unavailable | Unavailable | Unavailable | Estimated · #1/124 · not-comparable · 2 benchmarks |