Claude Fable 5 vs Qwen3.7 Max
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents
Wins 0 of 7 areas
—
Claude Fable 5 is the stronger all-rounder.Qwen3.7 Max is cheaper.
Scores updated · 45 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software90Claude Fable 59 of 9 tests
- ReasoningHard problems that need careful thinking40Claude Fable 54 of 5 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own40Claude Fable 54 of 4 tests
- MathCompetition and research-level math30Claude Fable 53 of 3 tests
- FactsGetting facts right instead of making them up21Claude Fable 52 of 3 tests
- Long documentsFinding answers in very long texts10Claude Fable 51 of 1 test
- Following instructionsDoing exactly what it is asked11Even1 each
Long documents rests on a single test.
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.
Qwen3.7 Max costs 83% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude Fable 5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+56points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+40.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+34.3points ahead
Where Qwen3.7 Max pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+38points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+17points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Fable 5
- SWE-bench ProClaude Fable 5 by 19.48060.6+19.4
- LiveBench · Agentic CodingClaude Fable 5 by 18.662.243.6+18.6
- SWE-bench VerifiedClaude Fable 5 by 14.69580.4+14.6
- Terminal-Bench HardClaude Fable 5 by 12.162.950.8+12.1
- LiveBench · CodingClaude Fable 5 by 11.88674.2+11.8
- SciCodeClaude Fable 5 by 11.56149.5+11.5
- LMArena · WebDevClaude Fable 5 by 109 rating points16241515+109 rating
- Terminal-Bench 2.1Claude Fable 5 by 10.184.674.5+10.1
- SWE-bench MultilingualClaude Fable 5 by 8.386.678.3+8.3
AgentsClaude Fable 5
- Terminal-Bench 4.0Claude Fable 5 by 40.942.41.5+40.9
- τ-Bench V3 · BankingClaude Fable 5 by 26.338.111.8+26.3
- GDPValClaude Fable 5 by 2455.631.6+24
- MCP AtlasClaude Fable 5 by 8.384.776.4+8.3
ReasoningClaude Fable 5
- CritPtClaude Fable 5 by 15.228.613.4+15.2
- Humanity's Last ExamClaude Fable 5 by 1555.540.5+15
- SimpleBenchClaude Fable 5 by 11.581.970.4+11.5
- LiveBench · ReasoningClaude Fable 5 by 6.489.783.3+6.4
- GPQA Diamondtie92.692.3tie
FactsClaude Fable 5
- AA-Omniscience · Non-hallucinationQwen3.7 Max by 3836.474.4+38
- AA-Omniscience · AccuracyClaude Fable 5 by 34.365.331.1+34.3
- SimpleQA VerifiedClaude Fable 5 by 14.970.755.8+14.9
MathClaude Fable 5
- FrontierMath Tier 4Claude Fable 5 by 5690.234.1+56
- FrontierMath Tiers 1-3 (v2)Claude Fable 5 by 22.48764.6+22.4
- LiveBench · MathematicsClaude Fable 5 by 10.79685.3+10.7
Following instructionsEven
- IFBenchQwen3.7 Max by 1763.580.5+17
- LiveBench · Instruction FollowingClaude Fable 5 by 1.875.874+1.8
Other results18 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.
- DeepSWEClaude Fable 5 by 527018+52
- DeepSWE 1.1Claude Fable 5 by 48.47021.6+48.4
- AA-OmniscienceClaude Fable 5 by 29.843.313.5+29.8
- Toolathlon VerifiedClaude Fable 5 by 28.277.949.7+28.2
- AndroidBenchClaude Fable 5 by 2884.556.5+28
- AA Agentic IndexClaude Fable 5 by 27.15123.9+27.1
- JobBenchClaude Fable 5 by 26.157.431.3+26.1
- PaperBenchClaude Fable 5 by 2488.864.8+24
- AA IntelligenceClaude Fable 5 by 20.149.629.5+20.1
- MLS Bench Litelower is betterQwen3.7 Max by 18.249.931.7+18.2
- livebench_languageClaude Fable 5 by 1190.779.7+11
- Artificial Analysis Coding IndexClaude Fable 5 by 10.576.566+10.5
- HLE (with tools)Claude Fable 5 by 10.363.853.5+10.3
- livebench_data_analysisClaude Fable 5 by 8.780.571.8+8.7
- Finance Agent v2Claude Fable 5 by 7.956.348.4+7.9
- WideSearchClaude Fable 5 by 681.275.2+6
- LiveBenchClaude Fable 5 by 478.374.3+4
- τ²-Bench Telecom (AA run)Claude Fable 5 by 3.898.594.7+3.8
Questions people ask
Which is better, Claude Fable 5 or Qwen3.7 Max?
Claude Fable 5 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and long documents. Qwen3.7 Max wins none, but costs 83% less. They are level on following instructions.
Which is better for coding?
Claude Fable 5. It wins 9 of the 9 coding tests both models report; Qwen3.7 Max wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Qwen3.7 Max costs $2.50 and $7.50. That makes Qwen3.7 Max about 83% cheaper for the same work.
How do you compare the two?
We use the 45 benchmark tests both models have published scores on. The verdict counts the 27 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.