Daily Arena
Match replay
Replaying a stored match — no models are called.
multilingual_support · roundTurn 0 / 3
The cheapest model that keeps up on quality appears here.
0 / 3
Claude Haiku 4.5
Anthropic
CL
€—score —
HP
100
Claude Opus 4.1
Anthropic
CL
€—score —
HP
100
Claude Sonnet 4.5
Anthropic
CL
€—score —
HP
100
Deep Research Preview (Apr-21-2026)
Google Gemini
DE
€—score —
HP
100
Deep Research Max Preview (Apr-21-2026)
Google Gemini
DE
€—score —
HP
100
DeepSeek v4 Pro
OpenRouter
DE
€—score —
HP
100
Customer
Press “Next turn” to begin.
Final verdict — cost, quality & voorsprong
| Players | Cost | Quality | Wins | Voorsprong / status |
|---|---|---|---|---|
| Claude Haiku 4.5 | €0.0053 | 96 | 1 | 100 HP |
| Claude Opus 4.1 | €0.0600 | 92 | 0 | 100 HP |
| Claude Sonnet 4.5 | €0.0132 | 91 | 0 | 100 HP |
| Deep Research Preview (Apr-21-2026) | €0.0000 | 0 | 0 | 82 HP |
| Deep Research Max Preview (Apr-21-2026) | €0.0000 | 0 | 0 | 100 HP |
| DeepSeek v4 Pro | €0.0008 | 85 | 0 | 100 HP |
0 / 3Drone damage = jury-majority strength · HP = live voorsprong · € = real cost
Honesty boundary
Advantage starts at 100; each turn the weakest active model loses the derived damage — damage = 16 + 24·margin, margin = (winner − runner-up) ÷ score-scale (deriveRoundOutcomes v8.1-tokonomix).
An exact tie has no decisive winner — no strike, no damage that turn.
Reaching 0 advantage is NOT elimination: every model still answers each turn. The real winner is the end-of-round judge panel below, shown for all models.
Damage reflects the relative gap between top scores, not absolute quality — winning a low-scoring turn deals the same as winning a high-scoring one.
Score-scale is the highest turn-score seen in this replay (0–10 or 0–100); one high turn can make the others look closer.
Zero model dispatch — pure render of the stored round. Switching the view changes the picture, never the numbers.