Coding Agents

Which Artificial Analysis coding agent and model pairing gives you the most intelligence per dollar — and where you are overpaying for the same score.

Best Models

The most cost-efficient model for each performance score, from all available models.

  1. Claude Code - Opus 5 (xhigh)
    68.1%
    $0.12/pt
  2. Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt

    70% cheaper · 4.1% worse than Claude Code - Opus 5 (xhigh)

  3. Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt

    15% cheaper · 2.4% worse than Grok Build - Grok 4.5 (high)

  4. Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt

    39% cheaper · 2.0% worse than Muse Code - Muse Spark 1.2 (xhigh)

  5. Codex - GPT-5.6 Luna (high)
    51.7%
    $0.02/pt

    28% cheaper · 7.9% worse than Opencode - Gemini 3.7 Flash (high)

  6. Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt

    87% cheaper · 1.9% worse than Codex - GPT-5.6 Luna (high)

  7. Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt

    29% cheaper · 11.5% worse than Codex - DeepSeek V4 Flash 0731 (max)

PerformanceCost ($1 blocks, 10¢ blocks)Empty
How it works

Models in the same performance bracket are treated as similar-score alternatives in the comparison.

Charts

Compare DeepSWE pass rate with mean cost. The top-right corner is the sweet spot: higher performance for less money.

55 selected models
Providers

Chart loads in the browser.

Comparison Table

55 models · sort by performance or price

Sort comparison table
  • Claude Code - Opus 5 (xhigh)· Anthropic
    68.1%
    $0.12/pt
    Best in class
  • Claude Code - Fable 5 (max) (with fallback)· Anthropic
    67.2%
    $0.17/pt
    Claude Code - Opus 5 (xhigh)
    68.1%
    $0.12/pt
    30% cheaper1.0% better

    43% pricier · 1.0% worse than Claude Code - Opus 5 (xhigh)

  • Claude Code - Opus 5 (max)· Anthropic
    67.0%
    $0.13/pt
    Claude Code - Opus 5 (xhigh)
    68.1%
    $0.12/pt
    9% cheaper1.1% better

    23% cheaper · 0.1% worse than Claude Code - Fable 5 (max) (with fallback)

  • Claude Code - Opus 5 (high)· Anthropic
    65.6%
    $0.06/pt
    Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt
    38% cheaper1.5% worse

    56% cheaper · 1.4% worse than Claude Code - Opus 5 (max)

  • Codex - GPT-5.6 Sol (max)· OpenAI
    65.1%
    $0.10/pt
    Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt
    62% cheaper1.0% worse

    64% pricier · 0.6% worse than Claude Code - Opus 5 (high)

  • Codex - GPT-5.6 Sol (high)· OpenAI
    64.1%
    $0.06/pt
    Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt
    37% cheaper0.0% worse

    40% cheaper · 0.9% worse than Codex - GPT-5.6 Sol (max)

  • Grok Build - Grok 4.5 (high)· xAI
    64.1%
    $0.04/pt
    Best in class

    37% cheaper · 0.0% worse than Codex - GPT-5.6 Sol (high)

  • Claude Code - Opus 5 (medium)· Anthropic
    64.1%
    $0.05/pt
    Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt
    23% cheaper0.0% better

    30% pricier · 0.0% worse than Grok Build - Grok 4.5 (high)

  • Codex - GPT-5.6 Sol (xhigh)· OpenAI
    63.3%
    $0.08/pt
    Grok Build - Grok 4.5 (high)
    64.1%
    $0.04/pt
    49% cheaper0.8% better

    51% pricier · 0.7% worse than Claude Code - Opus 5 (medium)

  • Kimi Code CLI - Kimi K3· Moonshot
    62.6%
    $0.05/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    33% cheaper1.0% worse

    36% cheaper · 0.7% worse than Codex - GPT-5.6 Sol (xhigh)

  • Claude Code - Opus 4.8 (max)· Anthropic
    62.1%
    $0.12/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    73% cheaper0.5% worse

    151% pricier · 0.5% worse than Kimi Code CLI - Kimi K3

  • Muse Code - Muse Spark 1.2 (xhigh)· Meta
    61.6%
    $0.03/pt
    Best in class

    73% cheaper · 0.5% worse than Claude Code - Opus 4.8 (max)

  • Codex - GPT-5.6 Sol (medium)· OpenAI
    61.6%
    $0.05/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    26% cheaper0.0% better

    36% pricier · 0.0% worse than Muse Code - Muse Spark 1.2 (xhigh)

  • Claude Code - Qwen3.8 Max· Alibaba
    61.3%
    $0.05/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    36% cheaper0.3% better

    15% pricier · 0.3% worse than Codex - GPT-5.6 Sol (medium)

  • Codex - GPT-5.5 (xhigh)· OpenAI
    61.0%
    $0.08/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    56% cheaper0.6% better

    47% pricier · 0.3% worse than Claude Code - Qwen3.8 Max

  • Codex - GPT-5.6 Terra (max)· OpenAI
    60.4%
    $0.04/pt
    Muse Code - Muse Spark 1.2 (xhigh)
    61.6%
    $0.03/pt
    23% cheaper1.2% better

    43% cheaper · 0.6% worse than Codex - GPT-5.5 (xhigh)

  • Opencode - Gemini 3.7 Flash (high)· Google
    59.6%
    $0.02/pt
    Best in class

    53% cheaper · 0.8% worse than Codex - GPT-5.6 Terra (max)

  • Claude Code - Opus 5 (low)· Anthropic
    59.4%
    $0.04/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    45% cheaper0.2% better

    81% pricier · 0.2% worse than Opencode - Gemini 3.7 Flash (high)

  • Claude Code - Opus 4.8 (xhigh)· Anthropic
    59.1%
    $0.10/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    78% cheaper0.6% better

    147% pricier · 0.3% worse than Claude Code - Opus 5 (low)

  • Opencode - Muse Spark 1.2 (xhigh)· Meta
    58.9%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    34% cheaper0.7% better

    66% cheaper · 0.2% worse than Claude Code - Opus 4.8 (xhigh)

  • Claude Code - Opus 4.8 (high)· Anthropic
    57.6%
    $0.07/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    66% cheaper2.1% better

    98% pricier · 1.3% worse than Opencode - Muse Spark 1.2 (xhigh)

  • Codex - GPT-5.6 Luna (max)· OpenAI
    57.2%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    16% cheaper2.5% better

    60% cheaper · 0.4% worse than Claude Code - Opus 4.8 (high)

  • Antigravity SDK - Gemini 3.7 Flash (high)· Google
    56.6%
    $0.02/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    9% cheaper3.1% better

    7% cheaper · 0.6% worse than Codex - GPT-5.6 Luna (max)

  • Codex - GPT-5.6 Terra (xhigh)· OpenAI
    56.0%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    33% cheaper3.6% better

    35% pricier · 0.5% worse than Antigravity SDK - Gemini 3.7 Flash (high)

  • Claude Code - Opus 4.8 (medium)· Anthropic
    55.8%
    $0.06/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    62% cheaper3.9% better

    76% pricier · 0.3% worse than Codex - GPT-5.6 Terra (xhigh)

  • Codex - GPT-5.5 (medium)· OpenAI
    55.3%
    $0.05/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    52% cheaper4.3% better

    20% cheaper · 0.5% worse than Claude Code - Opus 4.8 (medium)

  • Codex - GPT-5.6 Sol (low)· OpenAI
    55.2%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    24% cheaper4.4% better

    37% cheaper · 0.1% worse than Codex - GPT-5.5 (medium)

  • Opencode - Muse Spark 1.1 (xhigh)· Meta
    54.9%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    12% cheaper4.7% better

    14% cheaper · 0.3% worse than Codex - GPT-5.6 Sol (low)

  • Codex - GPT-5.6 Terra (high)· OpenAI
    54.6%
    $0.03/pt
    Opencode - Gemini 3.7 Flash (high)
    59.6%
    $0.02/pt
    19% cheaper5.0% better

    9% pricier · 0.3% worse than Opencode - Muse Spark 1.1 (xhigh)

  • Codex - GPT-5.6 Luna (xhigh)· OpenAI
    53.0%
    $0.02/pt
    Codex - GPT-5.6 Luna (high)
    51.7%
    $0.02/pt
    24% cheaper1.3% worse

    23% cheaper · 1.7% worse than Codex - GPT-5.6 Terra (high)

  • Codex - GPT-5.6 Luna (high)· OpenAI
    51.7%
    $0.02/pt
    Best in class

    24% cheaper · 1.3% worse than Codex - GPT-5.6 Luna (xhigh)

  • Claude Code - Opus 4.7 (max)· Anthropic
    51.6%
    $0.11/pt
    Codex - GPT-5.6 Luna (high)
    51.7%
    $0.02/pt
    84% cheaper0.1% better

    545% pricier · 0.1% worse than Codex - GPT-5.6 Luna (high)

  • Opencode - Opus 4.7 (medium)· Anthropic
    51.3%
    $0.06/pt
    Codex - GPT-5.6 Luna (high)
    51.7%
    $0.02/pt
    69% cheaper0.4% better

    50% cheaper · 0.3% worse than Claude Code - Opus 4.7 (max)

  • Codex - DeepSeek V4 Flash 0731 (max)· DeepSeek
    49.8%
    0.24¢/pt
    Best in class

    96% cheaper · 1.5% worse than Opencode - Opus 4.7 (medium)

  • Claude Code - Opus 4.8 (low)· Anthropic
    49.0%
    $0.04/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    95% cheaper0.7% better

    1734% pricier · 0.7% worse than Codex - DeepSeek V4 Flash 0731 (max)

  • Codex - GPT-5.6 Terra (medium)· OpenAI
    48.0%
    $0.02/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    87% cheaper1.7% better

    59% cheaper · 1.0% worse than Claude Code - Opus 4.8 (low)

  • Opencode - Gemini 3.6 Flash (high)· Google
    47.2%
    $0.04/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    94% cheaper2.5% better

    136% pricier · 0.8% worse than Codex - GPT-5.6 Terra (medium)

  • Cursor CLI - GPT-5.5 (medium)· xAI
    47.1%
    $0.04/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    94% cheaper2.7% better

    4% cheaper · 0.2% worse than Opencode - Gemini 3.6 Flash (high)

  • Cursor CLI - Opus 4.7 (medium)· xAI
    46.7%
    $0.06/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    96% cheaper3.1% better

    36% pricier · 0.4% worse than Cursor CLI - GPT-5.5 (medium)

  • Codex - GPT-5.6 Sol (none)· OpenAI
    43.4%
    $0.03/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    91% cheaper6.4% better

    49% cheaper · 3.3% worse than Cursor CLI - Opus 4.7 (medium)

  • Claude Code - GLM-5.2· Zhipu
    43.3%
    $0.15/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    98% cheaper6.4% better

    377% pricier · 0.1% worse than Codex - GPT-5.6 Sol (none)

  • Claude Code - Opus 4.7 (medium)· Anthropic
    42.4%
    $0.04/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    93% cheaper7.4% better

    73% cheaper · 0.9% worse than Claude Code - GLM-5.2

  • Codex - GPT-5.6 Luna (medium)· OpenAI
    42.0%
    $0.01/pt
    Codex - DeepSeek V4 Flash 0731 (max)
    49.8%
    0.24¢/pt
    73% cheaper7.8% better

    75% cheaper · 0.4% worse than Claude Code - Opus 4.7 (medium)

  • Claude Code - Sonnet 4.6 (medium)· Anthropic
    38.6%
    $0.05/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    96% cheaper0.3% worse

    361% pricier · 3.3% worse than Codex - GPT-5.6 Luna (medium)

  • Codex - GPT-5.6 Terra (low)· OpenAI
    38.6%
    $0.01/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    83% cheaper0.3% worse

    76% cheaper · 0.0% worse than Claude Code - Sonnet 4.6 (medium)

  • Cursor CLI - Composer 2.5· xAI
    38.3%
    0.22¢/pt
    Best in class

    83% cheaper · 0.3% worse than Codex - GPT-5.6 Terra (low)

  • Cursor CLI - Composer 2.5 Fast· xAI
    38.3%
    $0.01/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    85% cheaper0.0% better

    555% pricier · 0.0% better than Cursor CLI - Composer 2.5

  • Claude Code - Qwen3.7 Plus (thinking)· Alibaba
    38.1%
    $0.17/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    99% cheaper0.2% better

    1030% pricier · 0.2% worse than Cursor CLI - Composer 2.5 Fast

  • Claude Code - GLM-5.1· Zhipu
    36.9%
    $0.12/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    98% cheaper1.4% better

    32% cheaper · 1.2% worse than Claude Code - Qwen3.7 Plus (thinking)

  • Claude Code - Kimi K2.6· Moonshot
    33.9%
    $0.04/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    93% cheaper4.4% better

    72% cheaper · 3.0% worse than Claude Code - GLM-5.1

  • Gemini CLI - Gemini 3.1 Pro (high)· Google
    32.9%
    $0.03/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    92% cheaper5.4% better

    15% cheaper · 1.0% worse than Claude Code - Kimi K2.6

  • Claude Code - DeepSeek V4 Pro (high)· DeepSeek
    32.8%
    $0.02/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    83% cheaper5.5% better

    51% cheaper · 0.1% worse than Gemini CLI - Gemini 3.1 Pro (high)

  • Codex - GPT-5.6 Luna (low)· OpenAI
    25.1%
    0.77¢/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    56% cheaper13.2% better

    62% cheaper · 7.8% worse than Claude Code - DeepSeek V4 Pro (high)

  • Codex - GPT-5.6 Terra (none)· OpenAI
    23.1%
    $0.02/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    76% cheaper15.2% better

    88% pricier · 2.0% worse than Codex - GPT-5.6 Luna (low)

  • Codex - GPT-5.6 Luna (none)· OpenAI
    19.1%
    $0.02/pt
    Cursor CLI - Composer 2.5
    38.3%
    0.22¢/pt
    74% cheaper19.2% better

    9% cheaper · 4.0% worse than Codex - GPT-5.6 Terra (none)

PerformanceCost ($1 blocks, 10¢ blocks)Empty
How it works

Models in the same performance bracket are treated as similar-score alternatives in the comparison.

Model selector

Choose the model and reasoning effort rows to compare.

55 of 55 selected
ProvidersAll providers shown

Alibaba

Claude Code - Qwen3.8 Max

Alibaba

Claude Code - Qwen3.7 Plus

Alibaba

Anthropic

Claude Code - Opus 5

Anthropic

Claude Code - Fable 5 (with fallback)

Anthropic

Claude Code - Opus 4.8

Anthropic

Claude Code - Opus 4.7

Anthropic

Opencode - Opus 4.7

Anthropic

Claude Code - Sonnet 4.6

Anthropic

DeepSeek

Codex - DeepSeek V4 Flash 0731

DeepSeek

Claude Code - DeepSeek V4 Pro

DeepSeek

Google

Opencode - Gemini 3.7 Flash

Google

Antigravity SDK v0.1.8 - Gemini 3.7 Flash

Google

Opencode - Gemini 3.6 Flash

Google

Gemini CLI - Gemini 3.1 Pro

Google

Meta

Muse Code - Muse Spark 1.2

Meta

Opencode - Muse Spark 1.2

Meta

Opencode - Muse Spark 1.1

Meta

Moonshot

Kimi Code CLI - Kimi K3

Moonshot

Claude Code - Kimi K2.6

Moonshot

OpenAI

Codex - GPT-5.6 Sol

OpenAI

Codex - GPT-5.5

OpenAI

Codex - GPT-5.6 Terra

OpenAI

Codex - GPT-5.6 Luna

OpenAI

xAI

Grok Build - Grok 4.5

xAI

Cursor CLI - GPT-5.5

xAI

Cursor CLI - Opus 4.7

xAI

Cursor CLI - Composer 2.5

xAI

Cursor CLI - Composer 2.5 Fast

xAI

Zhipu

Claude Code - GLM-5.2

Zhipu

Claude Code - GLM-5.1

Zhipu