Model comparison
Claude Fable 5.1 vs Phi-4
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 31.2 on the Noometry Index. Phi-4 costs 229× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
MicrosoftSummary
- They share 20 benchmarks with published results for both. Claude Fable 5.1 scores higher in 9 categories and Phi-4 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5.1 leads 89.6 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5.1 and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Claude Fable 5.1 accepts more context: 1M tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5.1 | Phi-4 | |
|---|---|---|
| Provider | Anthropic | Microsoft |
| Noometry Index | 69.0 | 31.2 |
| Released | 2026-09-01 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | 1M | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $10 | $0.07 |
| Output $ / M tokens | $50 | $0.14 |
| Results tracked | 52 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5.1 leads
Claude Fable 5.1: 74.7 (#1), Phi-4: 34.4 (#239)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Coding | 1528 | 1231 |
| FrontierCode | 50.9% | — |
| CursorBench | 51.8% | — |
| LMArena WebDev | 1744 | — |
| FrontierSWE | 56.3% | — |
| SciCode | 63.1% | — |
| GSO | 88.2% | — |
| WeirdML | 92.9% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| MirrorCode | 73.3% | — |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 2,143 | — |
Agentic & Tool Use Claude Fable 5.1 leads
Claude Fable 5.1: 50.7 (#5), Phi-4: 22.8 (#128)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| APEX-Agents | 68.6% | — |
| Berkeley Function Calling Leaderboard | — | 28.8% |
| Remote Labor Index | 17.9% | — |
| BALROG | — | 11.6% |
| GDP.pdf | 29.6% | — |
| Vending-Bench 2 | 5,422 | — |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), Phi-4: 17.7 (#291)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| Chess Puzzles | 47% | 1% |
| LMArena Hard Prompts | 1526 | 1220 |
| Epoch Capabilities Index | 164.7 | 130.42 |
| ARC-AGI-2 | 90% | — |
| NYT Connections (extended) | 90% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 31.1% | — |
| EBR-Bench | 57.1% | — |
| LiveBench Reasoning | — | 47.8% |
| Mystery Game Puzzles | 58% | — |
| DTBench | 97.6% | — |
| LiveBench Data Analysis | — | 45.2% |
| LMCA | 65.5% | — |
| LiveBench | — | 41.6% |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), Phi-4: 20.8 (#285)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 13.8% |
| LMArena Math | 1525 | 1246 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| ProofBench | 100% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
| FrontierMath Erdős | 0% | — |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), Phi-4: 32.6 (#209)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Expert | 1535 | 1203 |
| GPQA Diamond | — | 56.1% |
| Humanity's Last Exam | 46.5% | — |
| SimpleQA Verified | 70.8% | — |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multimodal Not comparable
Claude Fable 5.1: 53.9 (#4), Phi-4: —
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Vision | 1318 | — |
| Blueprint-Bench 2 | 41.9% | — |
| Furniture Assembly | 70% | — |
| LMArena Document | 1513 | — |
Multilingual Claude Fable 5.1 leads
Claude Fable 5.1: 59.1 (#3), Phi-4: 37.2 (#237)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1507 | 1197 |
| LMArena Chinese | 1586 | 1212 |
| LMArena French | 1525 | 1224 |
| LMArena German | 1500 | 1222 |
| LMArena Japanese | 1543 | 1158 |
| LMArena Korean | 1534 | 1151 |
| LMArena Russian | 1521 | 1209 |
| LMArena Spanish | 1516 | 1234 |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), Phi-4: 60.4 (#251)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Claude Fable 5.1 leads
Claude Fable 5.1: 46.7 (#20), Phi-4: 36.9 (#226)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1522 | 1217 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), Phi-4: 40.5 (#244)
| Benchmark | Claude Fable 5.1 | Phi-4 |
|---|---|---|
| LMArena Text | 1510 | 1217 |
| LMArena Creative Writing | 1507 | 1182 |
| LMArena Multi-Turn | 1492 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 2162 | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Claude Fable 5.1 better than Phi-4?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 31.2 on the Noometry Index. Phi-4 costs 229× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Which is cheaper, Claude Fable 5.1 or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Claude Fable 5.1 lists at $10 and $50.
Is Claude Fable 5.1 or Phi-4 better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5.1 does, with 1M tokens against 128K.
How many benchmarks do Claude Fable 5.1 and Phi-4 share?
20 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Phi-4 has 37.