AutoBench Agronomy LLM Benchmark - December 2025

The first AutoBench run for the Agronomy domain with models Gemini 3 Pro, Gpt 5.1, Grok 4.1, Opus 4.5 and more

Past

Date

December 10, 2025

Version

2025-12-10

Models

New Models

Run data

Overall Comparison Costs Latency P99 Latency Domain

P99 latency breakdown across 10 specialized domains, measured in seconds. This 99th percentile metric reveals worst-case response times, helping you ensure consistent performance for production environments.

Model	Average (All Topics)	Coding	Creative Writing	Current News	General Culture	Grammar	History	Logics	Math	Science	Technology
Claude-opus-4.5	238.10s (#32)	-	-	-	-	-	-	-	-	-	-
Deepseek-v3.1	155.68s (#22)	-	-	-	-	-	-	-	-	-	-
GLM-4.5	200.73s (#29)	-	-	-	-	-	-	-	-	-	-
GLM-4.5-Air	144.36s (#20)	-	-	-	-	-	-	-	-	-	-
Kimi-K2-Instruct	86.47s (#10)	-	-	-	-	-	-	-	-	-	-
Kimi-K2-thinking	360.26s (#38)	-	-	-	-	-	-	-	-	-	-
Llama-3.1-nemotron-ultra-253b-v1	162.15s (#24)	-	-	-	-	-	-	-	-	-	-
Llama-4-scout	60.13s (#5)	-	-	-	-	-	-	-	-	-	-
Minimax-m2	238.55s (#33)	-	-	-	-	-	-	-	-	-	-
Mistral-large-2512	143.01s (#18)	-	-	-	-	-	-	-	-	-	-
Phi-3-mini-128k-instruct	142.96s (#17)	-	-	-	-	-	-	-	-	-	-
Qwen3-235B-A22B-Thinking-2507	254.75s (#34)	-	-	-	-	-	-	-	-	-	-
Qwen3-30B-A3B-Instruct-2507	174.47s (#26)	-	-	-	-	-	-	-	-	-	-
Qwen3-next-80b-a3b-thinking	126.90s (#16)	-	-	-	-	-	-	-	-	-	-
claude-3.5-haiku	73.19s (#8)	-	-	-	-	-	-	-	-	-	-
claude-haiku-4.5	365.38s (#39)	-	-	-	-	-	-	-	-	-	-
claude-sonnet-4.5	283.44s (#35)	-	-	-	-	-	-	-	-	-	-
deepSeek-R1-0528	159.18s (#23)	-	-	-	-	-	-	-	-	-	-
deepSeek-V3-0324	100.63s (#14)	-	-	-	-	-	-	-	-	-	-
deepSeek-v3.2-exp	381.40s (#40)	-	-	-	-	-	-	-	-	-	-
gemini-2.5-flash	90.11s (#12)	-	-	-	-	-	-	-	-	-	-
gemini-2.5-flash-lite	79.85s (#9)	-	-	-	-	-	-	-	-	-	-
gemini-2.5-pro	186.92s (#28)	-	-	-	-	-	-	-	-	-	-
gemini-3-pro-preview	143.02s (#19)	-	-	-	-	-	-	-	-	-	-
gemma-3-27b-it	111.58s (#15)	-	-	-	-	-	-	-	-	-	-
gpt-5	312.34s (#36)	-	-	-	-	-	-	-	-	-	-
gpt-5-mini	224.19s (#31)	-	-	-	-	-	-	-	-	-	-
gpt-5.1	347.66s (#37)	-	-	-	-	-	-	-	-	-	-
gpt-oss-120b	152.50s (#21)	-	-	-	-	-	-	-	-	-	-
grok-3-mini	97.02s (#13)	-	-	-	-	-	-	-	-	-	-
grok-4	219.87s (#30)	-	-	-	-	-	-	-	-	-	-
grok-4.1-fast	64.98s (#6)	-	-	-	-	-	-	-	-	-	-
grok-4.1-fast-thinking	176.91s (#27)	-	-	-	-	-	-	-	-	-	-
llama-3.3-nemotron-super-49b-v1.5	166.08s (#25)	-	-	-	-	-	-	-	-	-	-
llama-4-maverick	65.33s (#7)	-	-	-	-	-	-	-	-	-	-
magistral-small-2506	56.42s (#3)	-	-	-	-	-	-	-	-	-	-
nemotron-nano-9b-v2	88.99s (#11)	-	-	-	-	-	-	-	-	-	-
nova-lite-v1	41.75s (#1)	-	-	-	-	-	-	-	-	-	-
nova-pro-v1	45.94s (#2)	-	-	-	-	-	-	-	-	-	-
phi-4	59.91s (#4)	-	-	-	-	-	-	-	-	-	-