Model comparison
Claude Fable 5.1 vs Deepseek Coder v2
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 35.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. Claude Fable 5.1 scores higher in 8 categories and Deepseek Coder v2 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5.1 leads 89.6 to 34.9.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5.1 | Deepseek Coder v2 | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 69.0 | 35.9 |
| Released | 2026-09-01 | 2024-06-17 |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 128K | — |
| Input $ / M tokens | $10 | — |
| Output $ / M tokens | $50 | — |
| Results tracked | 52 | 24 |
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), Deepseek Coder v2: 38.1 (#183)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Coding | 1528 | 1251 |
| FrontierCode | 50.9% | — |
| CursorBench | 51.8% | — |
| LMArena WebDev | 1744 | — |
| FrontierSWE | 56.3% | — |
| SciCode | 63.1% | — |
| GSO | 88.2% | — |
| WeirdML | 92.9% | — |
| BigCodeBench Instruct | — | 48.2% |
| MirrorCode | 73.3% | — |
| BigCodeBench Complete | — | 59.7% |
| ALE-Bench | 2,143 | — |
| HumanEval+ | — | 82.3% |
| MBPP+ | — | 75.1% |
Agentic & Tool Use Not comparable
Claude Fable 5.1: 50.7 (#5), Deepseek Coder v2: —
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| APEX-Agents | 68.6% | — |
| Remote Labor Index | 17.9% | — |
| GDP.pdf | 29.6% | — |
| Vending-Bench 2 | 5,422 | — |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), Deepseek Coder v2: 23.6 (#176)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Hard Prompts | 1526 | 1207 |
| 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 | — |
| WinoGrande | — | 83.7% |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), Deepseek Coder v2: 34.9 (#190)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Math | 1525 | 1241 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 100% | — |
| FrontierMath Erdős | 0% | — |
| GSM8K | — | 94.5% |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), Deepseek Coder v2: 32.3 (#212)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Expert | 1535 | 1181 |
| Humanity's Last Exam | 46.5% | — |
| SimpleQA Verified | 70.8% | — |
| ARC (AI2) Challenge | — | 64.3% |
Multimodal Not comparable
Claude Fable 5.1: 53.9 (#4), Deepseek Coder v2: —
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| 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), Deepseek Coder v2: 36.3 (#240)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Non-English | 1507 | 1182 |
| LMArena Chinese | 1586 | 1201 |
| LMArena French | 1525 | 1185 |
| LMArena German | 1500 | 1164 |
| LMArena Japanese | 1543 | 1126 |
| LMArena Korean | 1534 | 1104 |
| LMArena Russian | 1521 | 1188 |
| LMArena Spanish | 1516 | 1153 |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), Deepseek Coder v2: 61.7 (#242)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1180 |
Long Context Claude Fable 5.1 leads
Claude Fable 5.1: 46.7 (#20), Deepseek Coder v2: 37.0 (#224)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Longer Query | 1522 | 1219 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), Deepseek Coder v2: 38.2 (#253)
| Benchmark | Claude Fable 5.1 | Deepseek Coder v2 |
|---|---|---|
| LMArena Text | 1510 | 1191 |
| LMArena Creative Writing | 1507 | 1120 |
| LMArena Multi-Turn | 1492 | 1177 |
| EQ-Bench Creative Writing | 2162 | — |
Frequently asked questions
Is Claude Fable 5.1 better than Deepseek Coder v2?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 35.9 on the Noometry Index.
Is Claude Fable 5.1 or Deepseek Coder v2 better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 38.1 in the Noometry coding category.
How many benchmarks do Claude Fable 5.1 and Deepseek Coder v2 share?
17 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Deepseek Coder v2 has 24.