Model comparison
Claude Fable 5 vs DeepSeek V4.1 Flash
Claude Fable 5 is the stronger model overall, scoring 66.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 76× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
DeepSeekSummary
- They share 37 benchmarks with published results for both. Claude Fable 5 scores higher in 10 categories and DeepSeek V4.1 Flash in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5 leads 76.8 to 50.2.
- The biggest single-benchmark swing is FrontierMath Tier 4: 90.2% for Claude Fable 5 and 26.8% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5 | DeepSeek V4.1 Flash | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 66.8 | 52.8 |
| Released | 2026-06-07 | 2026-09-09 |
| Weights | Proprietary | Open |
| Context window | 1M | 1M |
| Max output | 128K | 393K |
| Input $ / M tokens | $10 | $0.15 |
| Output $ / M tokens | $50 | $0.60 |
| Results tracked | 62 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5 leads
Claude Fable 5: 70.6 (#4), DeepSeek V4.1 Flash: 52.9 (#32)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena WebDev | 1625 | 1619 |
| SciCode | 61% | 51.9% |
| LMArena Coding | 1519 | 1506 |
| ALE-Bench | 2,041 | 1,092 |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| FrontierSWE | 47% | — |
| GSO | 78.4% | — |
| WeirdML | 91.9% | — |
| MirrorCode | 63.9% | — |
Agentic & Tool Use Claude Fable 5 leads
Claude Fable 5: 54.0 (#2), DeepSeek V4.1 Flash: 31.2 (#69)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | 63.6% | 39.5% |
| GDP.pdf | 30% | 19.8% |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| LMArena Search | 1230 | — |
| Vending-Bench 2 | 5,680 | — |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), DeepSeek V4.1 Flash: 50.2 (#36)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| NYT Connections (extended) | 92.7% | 89.6% |
| CritPt | 28.6% | 14.3% |
| LMArena Hard Prompts | 1508 | 1483 |
| Mystery Game Puzzles | 52% | 43% |
| DTBench | 98.4% | 89.9% |
| LMCA | 61.1% | 47% |
| Surface Evolver Bench | 95% | 46.3% |
| Epoch Capabilities Index | 162.06 | 154.9 |
| ARC-AGI-2 | 89.2% | — |
| SimpleBench | 81.9% | — |
| Kagi LLM Benchmark | 91.4% | — |
| ARC-AGI-1 | 98.5% | — |
| Chess Puzzles | 41% | — |
| EnigmaEval | 39.3% | — |
| EBR-Bench | 39.5% | — |
| Bench to the Future 3 | 0.13 | — |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), DeepSeek V4.1 Flash: 66.7 (#25)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 67.4% |
| FrontierMath Tier 4 | 90.2% | 26.8% |
| OTIS Mock AIME 2024-2025 | 100% | 98.3% |
| ProofBench | 95% | 54% |
| LMArena Math | 1519 | 1477 |
| FrontierMath Erdős | 0% | — |
Knowledge Claude Fable 5 leads
Claude Fable 5: 62.2 (#25), DeepSeek V4.1 Flash: 57.9 (#38)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 85.9% | 89.8% |
| LMArena Expert | 1534 | 1506 |
| SimpleQA Verified | 70.7% | — |
Multimodal Claude Fable 5 leads
Claude Fable 5: 45.3 (#17), DeepSeek V4.1 Flash: 39.1 (#61)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | 1324 | 1277 |
| Furniture Assembly | 35.8% | 34.2% |
| Blueprint-Bench 2 | 38.6% | — |
| LMArena Document | 1496 | — |
Multilingual Claude Fable 5 leads
Claude Fable 5: 57.3 (#9), DeepSeek V4.1 Flash: 55.0 (#35)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1481 | 1448 |
| LMArena Chinese | 1543 | 1497 |
| LMArena French | 1505 | 1452 |
| LMArena German | 1486 | 1484 |
| LMArena Japanese | 1506 | 1412 |
| LMArena Korean | 1488 | 1452 |
| LMArena Russian | 1504 | 1471 |
| LMArena Spanish | 1498 | 1459 |
Instruction Following Claude Fable 5 leads
Claude Fable 5: 78.6 (#8), DeepSeek V4.1 Flash: 77.3 (#26)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1502 | 1474 |
Long Context Claude Fable 5 leads
Claude Fable 5: 46.3 (#23), DeepSeek V4.1 Flash: 45.2 (#47)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1509 | 1475 |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), DeepSeek V4.1 Flash: 65.4 (#48)
| Benchmark | Claude Fable 5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1491 | 1462 |
| LMArena Creative Writing | 1494 | 1435 |
| EQ-Bench Creative Writing | 1943 | 1540 |
| LMArena Multi-Turn | 1504 | 1457 |
| EQ-Bench 4 | 1340 | — |
Frequently asked questions
Is Claude Fable 5 better than DeepSeek V4.1 Flash?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 52.8 on the Noometry Index. DeepSeek V4.1 Flash costs 76× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Which is cheaper, Claude Fable 5 or DeepSeek V4.1 Flash?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Claude Fable 5 lists at $10 and $50.
Is Claude Fable 5 or DeepSeek V4.1 Flash better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 52.9 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Fable 5 and DeepSeek V4.1 Flash share?
37 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and DeepSeek V4.1 Flash has 37.