Crypto x AI Daily

AI models and benchmark comparison

Overall intelligence scores of leading AI models in independent tests. Each family is shown at its strongest setting.

#ModelIntelligence indexContext
1Claude Opus 5.5 (max)Anthropic581M
2Claude Sonnet 5.5 (max)Anthropic561M
3Claude Fable 5.1 (max)Anthropic531M
3GPT-6 Astra (max)OpenAI531M
3Gemini 4 Argon (high)Google531M
6GPT-6.1 Sol (max)OpenAI521M
7Muse Spark 1.3 (max)Meta481M
8Grok 4.7 (xhigh)SpaceXAI46500k
8MiMo-V2.6-ProXiaomi461M
10Qwen3.8 Max (0902)Alibaba45984k
10GLM-5.3 (max)Z AI451M
12Kimi K3 (max)Kimi441.05M
12Step 5 PreviewStepFun441M
14Claude Haiku 5.5 (max)Anthropic431M
15Gemini 3.8 Flash (high)Google411M
16DeepSeek V4.1 Flash (max)DeepSeek391M
17MiniMax-M3MiniMax291M
18Nemotron 3 UltraNVIDIA23262k
19Mistral Medium 3.5Mistral14256k
19Nova 2.0 Pro Preview (medium)Amazon14256k
21gpt-oss-120b (high)OpenAI12131k

The intelligence index is Artificial Analysis’ independent score combining reasoning, knowledge, math and coding tests (higher is better). Context: how much text the model can read at once (tokens).

Source: Artificial Analysis · Updated: October 12, 2026

Takeaways

  • Highest score: Claude Opus 5.5 (max) (Anthropic) — 58
  • Best value: MiMo-V2.6-Pro — 46 / $0.13
  • Fastest: gpt-oss-120b (high) — 241 token/s

Full table with cost and speed

ModelCreatorIntelligence indexContextCost per taskSpeed (tokens/s)
Claude Opus 5.5 (max)Anthropic581M$5.9896
Claude Sonnet 5.5 (max)Anthropic561M$5.46137
Claude Fable 5.1 (max)Anthropic531M$7.6366
GPT-6 Astra (max)OpenAI531M$3.2663
Gemini 4 Argon (high)Google531M$1.99—
GPT-6.1 Sol (max)OpenAI521M$0.7266
Muse Spark 1.3 (max)Meta481M$1.60121
Grok 4.7 (xhigh)SpaceXAI46500k$3.7483
MiMo-V2.6-ProXiaomi461M$0.1348
Qwen3.8 Max (0902)Alibaba45984k$5.4136
GLM-5.3 (max)Z AI451M$2.0166
Kimi K3 (max)Kimi441.05M$2.0041
Step 5 PreviewStepFun441M$1.0389
Claude Haiku 5.5 (max)Anthropic431M$0.21235
Gemini 3.8 Flash (high)Google411M$1.24132
DeepSeek V4.1 Flash (max)DeepSeek391M$0.27208
MiniMax-M3MiniMax291M$0.51110
Nemotron 3 UltraNVIDIA23262k$0.60155
Mistral Medium 3.5Mistral14256k$0.50164
Nova 2.0 Pro Preview (medium)Amazon14256k—122
gpt-oss-120b (high)OpenAI12131k$0.11241

Other reasoning settings

ModelCreatorIntelligence indexContextCost per taskSpeed (tokens/s)
Claude Opus 5.5 (xhigh)Anthropic561M$3.4681
Claude Opus 5.5 (high)Anthropic541M$1.8282
Claude Opus 5.5 (medium)Anthropic511M$1.3482
Claude Fable 5.1 (xhigh)Anthropic531M$5.9862
Claude Fable 5.1 (high)Anthropic511M$3.9156
Claude Fable 5.1 (medium)Anthropic491M$2.9856
Claude Fable 5.1 (low)Anthropic471M$2.3757
Claude Sonnet 5.5 (xhigh)Anthropic521M$2.01106
Claude Sonnet 5.5 (high)Anthropic471M$0.88105
Claude Sonnet 5.5 (medium)Anthropic411M$0.48102
GPT-6 Astra (xhigh)OpenAI521M$2.3161
GPT-6 Astra (high)OpenAI511M$1.7356
GPT-6 Astra (medium)OpenAI501M$1.5455
GPT-6 Astra (low)OpenAI461M$0.8254
Muse Spark 1.3 (xhigh)Meta451M$1.37259
GPT-6.1 Sol (xhigh)OpenAI511M$0.3961
GPT-6.1 Sol (high)OpenAI501M$0.3258
GPT-6.1 Sol (medium)OpenAI481M$0.2155
GPT-6.1 Sol (low)OpenAI421M$0.1353
Grok 4.7 (high)SpaceXAI46500k$2.7382
Claude Haiku 5.5 (xhigh)Anthropic411M$0.12188
Claude Haiku 5.5 (high)Anthropic381M$0.08179

Single-benchmark highlights

BenchmarkModelScore
Humanity's Last ExamClaude Opus 5.5 (max)61.4%
Humanity's Last ExamClaude Fable 5.1 (max)59.1%
GPQA DiamondGPT-6 Astra (xhigh)96.3%
GPQA DiamondGPT-6 Astra (max)96.1%
GPQA DiamondGemini 3.8 Flash (high)95.3%

Source: Artificial Analysis · Updated: October 12, 2026