Skip to main content
ZICQ

Reasoning Rankings

ZICQ LLM rankings update daily across intelligence, coding, price, and context for selection and comparison.

Last updated: 41 min ago (2026-11-20 05:59)
Clear
Sources and methodology

声明支持 reasoning 推理 effort 的模型,按 Intelligence 指数排序。

📊 Evaluation Benchmarks & Methodology
1. Source:Model metadata and pricing are synced hourly from OpenRouter (openrouter.ai), covering commercial, open-weight, and preview/experimental versions.
2. Benchmarks:Intelligence, coding, and agentic indices are sourced directly from Artificial Analysis (artificialanalysis.ai) official benchmark indices.
3. Overall Score Weights:Overall Score = 60% × Intelligence Index + 25% × Coding Index + 15% × Agentic Index. Models without benchmark data are ranked at the end.
4. Referral links:"Try it" buttons marked "Referral" are referral links; ZICQ may earn a commission. Referral relationships never affect rankings or scores.
Statistics for this selection
Models tracked 479
Free & tunable 16
Vision 335
Tool calling 416
Reasoning 362
Max context xAI: Grok 4.20 Multi-Agent 2M tokens
Average known input price $1.56 / 1M USD / 1M tokens
Cheapest model inclusionAI: Ling 3.0 Tiny $0.015 / 1M
TOP by Intelligence Anthropic: Claude Fable 5.2 64.6
⚡ Side-by-Side Model PK & Token Cost Calculator Click "+ Compare" on any model below to swap · Live monthly cost calculator · Shareable comparison URL
Metric Anthropic: Claude Fable 5.2 Anthropic Anthropic: Claude Fable 5.2 (batch) Anthropic OpenAI: GPT-6.2 Astra OpenAI
Intelligence / Coding / Agentic 64.6 / 0.0 / 0.0 64.6 / 0.0 / 0.0 62.1 / 0.0 / 0.0
Context Window 1M tokens 1M tokens 1.1M tokens
Input / Output Price (1M) $8.00 / $40.00 $4.00 / $20.00 $8.00 / $40.00
Est. Monthly Cost (30d) Adjust inputs to calculate Adjust inputs to calculate Adjust inputs to calculate
  1. #81
    Thinking Machines: Inkling Small Thinkingmachines Audio-Visual Multimodal Benchmarked

    Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...

    Tools Vision Audio Reasoning
    Intelligence 12.9
    Context 524K Output ≤ 262K
    $0.45/1M $1.20/1M out cache $0.10/1M
    Solid Reasoning
  2. #82
    Thinking Machines: Inkling Small (batch) Thinkingmachines Audio-Visual Multimodal Benchmarked

    Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...

    Tools Vision Audio Reasoning
    Intelligence 12.9
    Context 524K Output ≤ 262K
    $0.36/1M $0.96/1M out cache $0.08/1M
    Solid Reasoning
  3. #83
    Thinking Machines: Inkling Thinkingmachines Audio-Visual Multimodal Benchmarked

    Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...

    Tools Vision Audio Reasoning
    Intelligence 12.5
    Context 524K Output ≤ 471K
    $1.00/1M $4.05/1M out cache $0.17/1M
    Solid Reasoning
  4. #84
    NVIDIA: Nemotron 3 Ultra NVIDIA General LLM Benchmarked

    NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

    Tools Reasoning Structured
    Intelligence 11.8
    Context 262K Output ≤ 16K
    $0.50/1M $2.20/1M out cache $0.10/1M
    Solid Reasoning
  5. #85
    NVIDIA: Nemotron 3 Ultra (free) NVIDIA General LLM Free Benchmarked

    NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

    Tools Reasoning
    Intelligence 11.8
    Context 1M Output ≤ 65K
    Free
    Free & Tunable
  6. #86
    Google: Gemini 3.5 Flash Lite Google DeepMind Audio-Video Multimodal Benchmarked

    Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.

    Tools Vision Audio Video Docs Reasoning Structured
    Intelligence 11.7
    Context 1.0M Output ≤ 65K
    $0.30/1M $2.50/1M out cache $0.03/1M
    Solid Reasoning
  7. #87
    Google: Gemini 3.5 Flash Lite (batch) Google DeepMind Audio-Video Multimodal Benchmarked

    Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.

    Tools Vision Audio Video Docs Reasoning Structured
    Intelligence 11.7
    Context 1.0M Output ≤ 65K
    $0.15/1M $1.25/1M out cache $0.01/1M
    Solid Reasoning
  8. #88
    Anthropic: Claude Haiku 4.5 Anthropic Doc & Vision Benchmarked

    Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...

    Tools Vision Docs Reasoning Structured
    Intelligence 7.4
    Context 200K Output ≤ 64K
    $1.00/1M $5.00/1M out cache $0.10/1M
    Solid Reasoning
  9. #89
    Anthropic: Claude Haiku 4.5 (batch) Anthropic Doc & Vision Benchmarked

    Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...

    Tools Vision Docs Reasoning Structured
    Intelligence 7.4
    Context 200K Output ≤ 64K
    $0.50/1M $2.50/1M out cache $0.05/1M
    Solid Reasoning
  10. #90
    Meta: Muse Glimmer 30B Meta 视觉多模态 Benchmarked

    Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...

    Tools Vision Reasoning Structured
    Intelligence 6.3
    Context 131K Output ≤ 117K
    $0.35/1M $1.50/1M out cache $0.04/1M
    Solid Reasoning
  11. #91
    Meta: Muse Glimmer 30B (batch) Meta 视觉多模态 Benchmarked

    Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...

    Tools Vision Reasoning Structured
    Intelligence 6.3
    Context 131K Output ≤ 16K
    $0.24/1M $0.96/1M out cache $0.03/1M
    Solid Reasoning
  12. #92
    Cohere: Command A+ Cohere 视觉多模态 Benchmarked

    Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K context window, supports native tool calling with strict tool schemas, structured...

    Tools Vision Reasoning Structured
    Intelligence 6.2
    Context 192K Output ≤ 64K
    $0.30/1M $1.50/1M out cache $0.15/1M
    Solid Reasoning
  13. #93
    Google: Gemini 3.1 Flash Lite Preview Google DeepMind Audio-Video Multimodal Benchmarked

    Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases. It outperforms Gemini 2.5 Flash Lite on overall quality and approaches Gemini 2.5 Flash performance across...

    Tools Vision Audio Video Docs Reasoning Structured
    Intelligence 6.2
    Context 1.0M Output ≤ 65K
    $0.25/1M $1.50/1M out cache $0.02/1M
    Solid Reasoning
  14. #94
    LiquidAI: LFM3-1.7B Liquid General LLM Benchmarked

    A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prompts and outputs may be retained and used...

    Tools Reasoning Structured
    Intelligence 6.0
    Context 262K Output ≤ 8K
    $0.08/1M $0.20/1M out cache $0.02/1M
    Solid Reasoning
  15. #95
    LiquidAI: LFM3-1.7B (free) Liquid General LLM Free Benchmarked

    A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prompts and outputs may be retained and used...

    Tools Reasoning Structured
    Intelligence 6.0
    Context 262K Output ≤ 8K
    Free
    Free & Tunable
  16. #96
    NVIDIA: Nemotron 3.5 Lightning NVIDIA General LLM Benchmarked

    NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...

    Tools Reasoning Structured
    Intelligence 5.5
    Context 1.0M Output ≤ 235K
    $0.07/1M $0.20/1M out cache $0.04/1M
    Solid Reasoning
  17. #97
    NVIDIA: Nemotron 3.5 Lightning (free) NVIDIA General LLM Free Benchmarked

    NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...

    Tools Reasoning
    Intelligence 5.5
    Context 1M Output ≤ 65K
    Free
    Free & Tunable
  18. #98
    IBM: Granite 4.2 8B Ibm-Granite General LLM Benchmarked

    Granite 4.2 8B is a dense reasoning model from IBM. It is suited for mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. It supports full, low-effort,...

    Tools Reasoning Structured
    Intelligence 5.1
    Context 131K Output ≤ 117K
    $0.06/1M $0.15/1M out cache $0.01/1M
    Solid Reasoning
  19. #99
    Mistral: Mistral Medium 3.5 Mistral AI Doc & Vision Benchmarked

    Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...

    Tools Vision Docs Reasoning Structured
    Intelligence 4.6
    Context 262K Output ≤ 209K
    $1.50/1M $7.50/1M out
    Solid Reasoning
  20. #100
    Mistral: Mistral Medium 3.5 (batch) Mistral AI Doc & Vision Benchmarked

    Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...

    Tools Vision Docs Reasoning Structured
    Intelligence 4.6
    Context 262K Output ≤ 209K
    $0.75/1M $3.75/1M out
    Solid Reasoning