Best LLM for game development (2026)
Which model handles game code and mechanics best?
Rebuilt from live model data · August 3, 2026
How we ranked
Ranked by Design Arena position in the "gamedev" category — head-to-head win rates across scored models, reported through the OpenRouter model API.
Top 3 for game development
- 1
Claude Fable 5
Win rate: 66.5% $10.00 in / $50.00 out per 1M 1M context 4 providersClaude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...
Vision Tool calling Reasoning Released Jun 9, 2026 - 2
GLM 5.2
Win rate: 62% $0.741 in / $2.33 out per 1M 1.0M context 29 providersGLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...
Open weights Tool calling Reasoning Released Jun 16, 2026 - 3
Claude Sonnet 5
Win rate: 59.9% $2.00 in / $10.00 out per 1M 1M context 4 providersSonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
Vision Tool calling Reasoning Released Jun 30, 2026
Full ranking
| # | Model | Win rate | Output / 1M | Context | Providers | Weights |
|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 | 66.5% | $50.00 | 1M | 4 | Closed |
| 2 | GLM 5.2 | 62% | $2.33 | 1.0M | 29 | Open |
| 3 | Claude Sonnet 5 | 59.9% | $10.00 | 1M | 4 | Closed |
| 4 | GPT-5.5 | 61.3% | $30.00 | 1.1M | 2 | Closed |
| 5 | MiMo-V2.5-Pro | 62% | $0.696 | 1.1M | 6 | Open |
| 6 | Claude Opus 4.6 | 62.9% | $25.00 | 1M | 4 | Closed |
| 7 | Claude Opus 4.7 | 61% | $25.00 | 1M | 4 | Closed |
| 8 | Grok 4.5 | 54.2% | $6.00 | 500K | 1 | Closed |
| 9 | GLM 5.1 | 61.2% | $2.99 | 205K | 19 | Open |
| 10 | Muse Spark 1.1 | 55.6% | $4.25 | 1.0M | 1 | Closed |
| 11 | Qwen3.7 Max | 59.1% | $4.42 | 1M | 1 | Closed |
| 12 | Gemini 3.5 Flash | 57.6% | $9.00 | 1.0M | 2 | Closed |
FAQ
Which model handles game code and mechanics best?
Claude Fable 5 leads for game development with win rate 66.5%, at $50.00 per 1M output tokens. GLM 5.2 (62%) and Claude Sonnet 5 (59.9%) follow. Ranked by Design Arena position in the "gamedev" category — head-to-head win rates across scored models, reported through the OpenRouter model API.
How is this ranking produced?
Ranked by Design Arena position in the "gamedev" category — head-to-head win rates across scored models, reported through the OpenRouter model API. The list rebuilds from the live model catalogue on every deploy, so a model that launches or changes price appears here without an editor rewriting the page. Last rebuild: August 3, 2026.
What does the top pick cost to run?
Claude Fable 5 costs about $25.00 for a workload of 1M input + 300K output tokens, based on the cheapest of 4 provider(s).