Claude Fable 5 vs DeepSeek-V4.1-Flash
Wins 6 of 8 areas
Agents · Reasoning · Facts · Images and charts · Math · Following instructions
Wins 1 of 8 areas
Long documents
Claude Fable 5 is the stronger all-rounder.DeepSeek-V4.1-Flash is cheaper and better at long documents.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking60Claude Fable 56 of 6 tests
- FactsGetting facts right instead of making them up30Claude Fable 53 of 3 tests
- MathCompetition and research-level math30Claude Fable 53 of 3 tests
- Images and chartsUnderstanding pictures, charts and video20Claude Fable 52 of 2 tests
- AgentsCarrying out multi-step tasks on its own10Claude Fable 51 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked10Claude Fable 51 of 1 test
- Long documentsFinding answers in very long texts01DeepSeek-V4.1-Flash1 of 1 test
- CodingWriting and fixing software22Even2 each · 1 tie
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
DeepSeek-V4.1-Flash costs 98% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Fable 5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+43.9points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+32.9points ahead
- Short factual questions, answered correctlySimpleQA Verified+27.8points ahead
Where DeepSeek-V4.1-Flash pulls ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+15.1points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+6points ahead
- Reasons across sets of long documentsAA-LCR+1.7points ahead
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- LiveBench · Agentic CodingDeepSeek-V4.1-Flash by 15.162.277.3+15.1
- SciCodeClaude Fable 5 by 9.16151.9+9.1
- LiveBench · CodingClaude Fable 5 by 68680+6
- Terminal-Bench 2.1DeepSeek-V4.1-Flash by 684.690.6+6
- LMArena · WebDevtie16241619tie
AgentsClaude Fable 5
- Terminal-Bench 4.0Claude Fable 5 by 15.642.426.8+15.6
- GDPValtie55.555tie
ReasoningClaude Fable 5
- Humanity's Last ExamClaude Fable 5 by 16.355.539.2+16.3
- SimpleBenchClaude Fable 5 by 15.281.966.7+15.2
- CritPtClaude Fable 5 by 14.328.614.3+14.3
- ARC-AGI-2Claude Fable 5 by 11.189.278.1+11.1
- LiveBench · ReasoningClaude Fable 5 by 389.786.7+3
- GPQA DiamondClaude Fable 5 by 1.792.690.9+1.7
FactsClaude Fable 5
- AA-Omniscience · Non-hallucinationClaude Fable 5 by 32.936.43.5+32.9
- SimpleQA VerifiedClaude Fable 5 by 27.870.742.9+27.8
- AA-Omniscience · AccuracyClaude Fable 5 by 1965.346.4+19
Images and chartsClaude Fable 5
- MMMU-ProClaude Fable 5 by 7.284.277+7.2
- LMArena · VisionClaude Fable 5 by 47 rating points13251278+47 rating
MathClaude Fable 5
- FrontierMath Tier 4Claude Fable 5 by 43.990.246.3+43.9
- FrontierMath Tiers 1-3 (v2)Claude Fable 5 by 15.88771.2+15.8
- LiveBench · MathematicsClaude Fable 5 by 2.79693.3+2.7
Long documentsDeepSeek-V4.1-Flash
- AA-LCRDeepSeek-V4.1-Flash by 1.782.384+1.7
Following instructionsClaude Fable 5
- LiveBench · Instruction FollowingClaude Fable 5 by 5.875.870+5.8
Other results14 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Program BenchClaude Fable 5 by 6686.320.3+66
- AA-OmniscienceClaude Fable 5 by 48.643.3-5.3+48.6
- Terminal-Bench 4.0Claude Fable 5 by 13.344.531.2+13.3
- GDP (Surge AI)Claude Fable 5 by 12.629.817.2+12.6
- OSWorld 2.0Claude Fable 5 by 11.866.154.3+11.8
- AA IntelligenceClaude Fable 5 by 10.149.639.5+10.1
- livebench_languageClaude Fable 5 by 9.590.781.2+9.5
- Terminal-Bench-Science 0.1Claude Fable 5 by 924.715.7+9
- Agents' Last ExamDeepSeek-V4.1-Flash by 6.125.731.8+6.1
- ARC-AGI-1Claude Fable 5 by 598.593.5+5
- DeepSWE 1.1DeepSeek-V4.1-Flash by 4.27074.2+4.2
- charxiv_rq_with_pythonClaude Fable 5 by 3.889.485.6+3.8
- livebench_data_analysisClaude Fable 5 by 1.280.579.3+1.2
- HLE (with tools)tie63.863.9tie
Questions people ask
Which is better, Claude Fable 5 or DeepSeek-V4.1-Flash?
Claude Fable 5 wins six of the eight areas we test: agents, reasoning, facts, images and charts, math and following instructions. DeepSeek-V4.1-Flash wins long documents, and costs 98% less. They are level on coding.
Which is better for coding?
Neither. They win 2 coding tests each of the 5 both models report, and 1 is a tie.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; DeepSeek-V4.1-Flash costs $0.30 and $1.20. That makes DeepSeek-V4.1-Flash about 98% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.