Best LLMs for Game Development — Anthropic
The best Anthropic model for game development is Claude Fable 5, rated 1399 on Design Arena's game-dev board for playable games and game mechanics. Claude Sonnet 5 (1339) and Claude Opus 4.7 (1327) round out the top three.
- 1
claude-fable-5ReasoningToolsJSON+159.9 intel · $10.00/M · 64 t/s1399Elo - 2claude-sonnet-5ReasoningToolsJSON+153.4 intel · $2.00/M · 79 t/s1339Elo
- 3claude-opus-4.7ReasoningToolsJSON+153.5 intel · $5.00/M · 1M ctx1327Elo
- 4claude-opus-4.6ReasoningToolsJSON+137.8 intel · $5.00/M · 1M ctx1324Elo
- 5claude-sonnet-4.6ReasoningToolsJSON+135.9 intel · $3.00/M · 1M ctx1304Elo
- 6claude-opus-4.8ReasoningToolsJSON+155.7 intel · $5.00/M · 77 t/s1290Elo
- 7claude-opus-4.5ReasoningToolsJSON+134.7 intel · $5.00/M · 44 t/s1275Elo
- 8claude-opus-4ReasoningToolsVision25.5 intel · $15.00/M · 22 t/s1215Elo
- 9claude-sonnet-4.5ReasoningToolsJSON+129.3 intel · $3.00/M · 38 t/s1211Elo
- 10claude-sonnet-4ReasoningToolsVision25.5 intel · $3.00/M · 50 t/s1184Elo
- 11claude-haiku-4.5ReasoningToolsJSON+123.7 intel · $1.00/M · 94 t/s1142Elo
How this is ranked
AI models ranked by Design Arena Elo in the game-development category — building playable games and game mechanics, judged head-to-head by humans. Games are the hardest kind of generated code to fake: the result either runs and plays or it doesn't, which makes this ranking unusually honest about a model's ability to hold state, physics and a game loop together.
Frequently asked
What is the best Anthropic model for game development?
The best Anthropic model for game development is Claude Fable 5, rated 1399 on Design Arena's game-dev board for playable games and game mechanics. Claude Sonnet 5 (1339) and Claude Opus 4.7 (1327) round out the top three.
What's a good alternative to Claude Fable 5?
Claude Sonnet 5 (1339) is the closest alternative on this metric, followed by Claude Opus 4.7 (1327). See the full ranking above for the tradeoffs.
How many Anthropic models are there?
modelgrep tracks 17 Anthropic models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Claude Opus 5. 11 of them qualify for this ranking.