Benchmark · agentic

GDPval-AA (Elo)

14 results 10 models

Artificial Analysis's pairwise-graded Elo ranking on real-world economically valuable knowledge work; four-digit scores (~640-1950 Elo), so never file a percentage here. Not Chatbot Arena Elo (lmsys-arena) and not OpenAI's percentage GDPval.

Gemini 3.6 Flash Gemini 3.5 Flash-Lite DeepSeek-V4-Flash-0731 Grok 4.6 Gemini 3.7 Flash GLM-5.3 GLM-5.3-Flash Claude Fable 5.1 Muse Spark 1.3 K2-Horizon-375B-A23B

Timeline

Date	Model	Score	Source
2026-09-07	K2-Horizon-375B-A23B	1441.0Elo	IFM releases K2 Horizon: six Apache 2.0 models from 0.9B to 375B
2026-09-04	Muse Spark 1.3	1754.0Elo	Meta releases Muse Spark 1.3 with fewer tool calls and tokens
2026-09-01	Claude Fable 5.1	1853.0Elo	Anthropic releases Claude Fable 5.1 with 52.6% on Terminal-Bench-Science and cheaper cache reads
2026-08-27	GLM-5.3-Flash	1770.0Elo	NVIDIA buys HuggingFace for $13B; Z.ai launches GLM-5.3-Flash
2026-08-27	GLM-5.3-Flash	1770.0Elo	Z.ai launches GLM-5.3-Flash, a 320B-parameter multimodal model with 1M context
2026-08-21	Grok 4.6	1577.0Elo	SpaceXAI introduces Grok 4.6 with Cursor data for agentic work
2026-08-17	GLM-5.3	1769.0Elo	Z.ai launches GLM-5.3 coding model; Cursor joins SpaceXAI
2026-08-14	Gemini 3.7 Flash	1525.0Elo	Google releases Gemini 3.7 Flash with algorithmic improvements and lower pricing
2026-08-13	Grok 4.6	1753.0Elo	SpaceXAI releases Grok 4.6 with 500K context and improved agentic reasoning
2026-08-13	Grok 4.6	1753.0Elo	xAI releases Grok 4.6, Alibaba drops Qwen3.8-Max, DeepSeek V4 Pro GA
2026-08-13	Grok 4.6	1753.0Elo	xAI releases Grok 4.6; Alibaba opens Qwen3.8-MoE; DeepSeek V4 Pro GA
2026-08-07	DeepSeek-V4-Flash-0731	1558.0Elo	DeepSeek releases V4-Flash-0731, outperforming V4-Pro at lower cost
2026-07-27	Gemini 3.6 Flash	1421.0Elo	Google releases Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
2026-07-27	Gemini 3.5 Flash-Lite	1140.0Elo	Google releases Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber