VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

Claude Opus 4.8 vs Gemini 3.6 Flash

AnthropicvsGoogle40 shared benchmarks319 head-to-head
BenchmarkClaude Opus 4.8Gemini 3.6 Flash
AA Agentic Index49.440.5
AA Intelligence57.352
AA-LCR7379
AA-Omniscience28.822.1
aime_202610096.7
arc_agi_192.534.5
ARC-AGI-272.12.6
arena_elo14821481
arena_text_factuality14641470
Artificial Analysis Coding Index74.369.2
charxiv_reasoning_with_tools89.989.4
charxiv_rq80.585.2
coding_arena_elo15391538
critpt20.910.6
DeepSWE5949
DeepSWE 1.15949
GDP (Surge AI)84.822
gdpval54.246.1
GDPval-AA v215931421
GPQA Diamond93.692.8
HLE57.940.8
hmmt_feb_202696.789.4
livebench_agentic_coding50.543.4
livebench_coding81.877.9
livebench_data_analysis6663
livebench_instruction_following7275.4
livebench_language79.783.9
livebench_math94.386.4
livebench_reasoning89.285.2
matharena_visual_math_overall81.687.8
MMMU-Pro78.983.2
OmniScience Accuracy48.850
OmniScience Non-Hallucination60.744.4
OSWorld 2.055.733.8
OSWorld-Verified83.483
scicode53.552.7
simpleqa_verified39.568.7
SWE-bench Pro69.258.7
TauBench V3 - Banking34.229.9
Terminal-Bench 2.18578

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.