Claude Fable 5 vs Hy3-preview
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
Claude Fable 5 is the stronger all-rounder.Hy3-preview is cheaper.
Scores updated · 29 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 software70Claude Fable 57 of 7 tests
- AgentsCarrying out multi-step tasks on its own60Claude Fable 56 of 6 tests
- ReasoningHard problems that need careful thinking30Claude Fable 53 of 3 tests
- FactsGetting facts right instead of making them up20Claude Fable 52 of 2 tests
- Long documentsFinding answers in very long texts10Claude Fable 51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Fable 51 of 1 test
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.
Hy3-preview costs 99% 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
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+41.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+33.4points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+31.1points ahead
Where Hy3-preview pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 29 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Fable 5
- Terminal-Bench HardClaude Fable 5 by 31.162.931.8+31.1
- SWE-bench ProClaude Fable 5 by 22.18057.9+22.1
- Terminal-Bench 2.1Claude Fable 5 by 20.284.664.4+20.2
- SWE-bench VerifiedClaude Fable 5 by 179578+17
- SciCodeClaude Fable 5 by 12.46148.6+12.4
- LMArena · WebDevClaude Fable 5 by 116 rating points16241508+116 rating
- SWE-bench MultilingualClaude Fable 5 by 10.886.675.8+10.8
AgentsClaude Fable 5
- Terminal-Bench 4.0Claude Fable 5 by 41.942.40.5+41.9
- GDPValClaude Fable 5 by 27.355.628.3+27.3
- AA ApexAgentsClaude Fable 5 by 17.743.325.6+17.7
- τ-Bench V3 · BankingClaude Fable 5 by 15.238.122.9+15.2
- MCP AtlasClaude Fable 5 by 5.684.779.1+5.6
- BrowseCompClaude Fable 5 by 3.287.484.2+3.2
ReasoningClaude Fable 5
- CritPtClaude Fable 5 by 23.728.64.9+23.7
- Humanity's Last ExamClaude Fable 5 by 2255.533.5+22
- GPQA DiamondClaude Fable 5 by 2.992.689.7+2.9
FactsClaude Fable 5
- AA-Omniscience · AccuracyClaude Fable 5 by 33.465.332+33.4
- AA-Omniscience · Non-hallucinationClaude Fable 5 by 10.536.425.9+10.5
Following instructionsClaude Fable 5
- IFBenchClaude Fable 5 by 15.563.548+15.5
Other results9 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.
- AA-OmniscienceClaude Fable 5 by 61.843.3-18.5+61.8
- DeepSWEClaude Fable 5 by 427028+42
- τ²-Bench Telecom (AA run)Claude Fable 5 by 3198.567.5+31
- AA Agentic IndexClaude Fable 5 by 25.45125.6+25.4
- AA IntelligenceClaude Fable 5 by 24.349.625.3+24.3
- Artificial Analysis Coding IndexClaude Fable 5 by 17.776.558.8+17.7
- ToolathlonClaude Fable 5 by 13.261.748.5+13.2
- WideSearchClaude Fable 5 by 4.881.276.4+4.8
- DeepSearchQA (F1)Claude Fable 5 by 3.294.291+3.2
Questions people ask
Which is better, Claude Fable 5 or Hy3-preview?
Claude Fable 5 wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Hy3-preview wins none, but costs 99% less.
Which is better for coding?
Claude Fable 5. It wins 7 of the 7 coding tests both models report; Hy3-preview wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Hy3-preview costs $0.14 and $0.56. That makes Hy3-preview about 99% cheaper for the same work.
How do you compare the two?
We use the 29 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 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.