Model comparison
Claude Fable 5.1 vs Llama 3.2 3B
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 28.9 on the Noometry Index. Llama 3.2 3B costs 167× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
MetaSummary
- They share 14 benchmarks with published results for both. Claude Fable 5.1 scores higher in 9 categories and Llama 3.2 3B 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 32.4.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Claude Fable 5.1 accepts more context: 1M tokens versus 131K.
- Llama 3.2 3B has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5.1 | Llama 3.2 3B | |
|---|---|---|
| Provider | Anthropic | Meta |
| Noometry Index | 69.0 | 28.9 |
| Released | 2026-09-01 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 1M | 131K |
| Max output | 128K | 118K |
| Input $ / M tokens | $10 | $0.05 |
| Output $ / M tokens | $50 | $0.33 |
| Results tracked | 52 | 18 |
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), Llama 3.2 3B: 27.6 (#319)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Coding | 1528 | 1098 |
| FrontierCode | 50.9% | — |
| CursorBench | 51.8% | — |
| LMArena WebDev | 1744 | — |
| FrontierSWE | 56.3% | — |
| SciCode | 63.1% | — |
| GSO | 88.2% | — |
| WeirdML | 92.9% | — |
| BigCodeBench Instruct | — | 23.4% |
| MirrorCode | 73.3% | — |
| BigCodeBench Complete | — | 28.3% |
| ALE-Bench | 2,143 | — |
Agentic & Tool Use Claude Fable 5.1 leads
Claude Fable 5.1: 50.7 (#5), Llama 3.2 3B: 20.1 (#143)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| APEX-Agents | 68.6% | — |
| Berkeley Function Calling Leaderboard | — | 21.9% |
| Remote Labor Index | 17.9% | — |
| BALROG | — | 10.1% |
| GDP.pdf | 29.6% | — |
| Vending-Bench 2 | 5,422 | — |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), Llama 3.2 3B: 21.0 (#228)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Hard Prompts | 1526 | 1095 |
| ARC-AGI-2 | 90% | — |
| NYT Connections (extended) | 90% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 31.1% | — |
| Chess Puzzles | 47% | — |
| EBR-Bench | 57.1% | — |
| Mystery Game Puzzles | 58% | — |
| DTBench | 97.6% | — |
| LMCA | 65.5% | — |
| Epoch Capabilities Index | 164.7 | — |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), Llama 3.2 3B: 32.4 (#214)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Math | 1525 | 1126 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 100% | — |
| FrontierMath Erdős | 0% | — |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), Llama 3.2 3B: 29.7 (#235)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Expert | 1535 | 1090 |
| Humanity's Last Exam | 46.5% | — |
| SimpleQA Verified | 70.8% | — |
Multimodal Not comparable
Claude Fable 5.1: 53.9 (#4), Llama 3.2 3B: —
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| 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), Llama 3.2 3B: 26.2 (#281)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Non-English | 1507 | 1019 |
| LMArena Chinese | 1586 | 1017 |
| LMArena German | 1500 | 1056 |
| LMArena Russian | 1521 | 949 |
| LMArena French | 1525 | — |
| LMArena Japanese | 1543 | — |
| LMArena Korean | 1534 | — |
| LMArena Spanish | 1516 | — |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), Llama 3.2 3B: 56.0 (#275)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Instruction Following | 1517 | 1089 |
Long Context Claude Fable 5.1 leads
Claude Fable 5.1: 46.7 (#20), Llama 3.2 3B: 33.4 (#261)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Longer Query | 1522 | 1100 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), Llama 3.2 3B: 24.7 (#307)
| Benchmark | Claude Fable 5.1 | Llama 3.2 3B |
|---|---|---|
| LMArena Text | 1510 | 1110 |
| LMArena Creative Writing | 1507 | 1094 |
| EQ-Bench Creative Writing | 2162 | 595 |
| LMArena Multi-Turn | 1492 | 1105 |
Frequently asked questions
Is Claude Fable 5.1 better than Llama 3.2 3B?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 28.9 on the Noometry Index. Llama 3.2 3B costs 167× 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 Llama 3.2 3B?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Claude Fable 5.1 lists at $10 and $50.
Is Claude Fable 5.1 or Llama 3.2 3B better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5.1 does, with 1M tokens against 131K.
How many benchmarks do Claude Fable 5.1 and Llama 3.2 3B share?
14 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Llama 3.2 3B has 18.