All General Coding Reasoning Multimodal Safety Specialized Agentic Language Long Context Speed Text-to-Image Image Editing Text-to-Video Image-to-Video
Artificial Analysis Coding Index259 Chatbot Arena Coding278 CodeElo HumanEval LiveCodeBench343 MBPP SWE-bench SWE-bench Verified SciCode229

Coding

Composite performance across 4 benchmarks · Normalized 0–100 · At least 2 scores required · Ranked by average performance · Top 50 of 239 eligible models

# Model Provider AA Coding Arena Coding LiveCodeBench SciCode Avg Coverage Price Value
1 Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) Anthropic 81.6 — — 63.1%
96.9
2/4 $60.00/M 1.6
2 Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) Anthropic 80.7 — — 60.9%
94.5
2/4 $60.00/M 1.6
3 Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) Anthropic 76.5 — — 61.0%
92.0
2/4 $60.00/M 1.5
4 Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) Anthropic 79.1 — — 58.7%
91.6
2/4 $60.00/M 1.5
5 Muse Spark 1.3 (xhigh) Meta 76.5 — — 59.7%
90.9
2/4 $5.50/M 16.5
6 Kimi K3 (max) Moonshot AI 76.2 — — 59.5%
90.6
2/4 $15.50/M 5.8
7 GPT-5.6 Sol (xhigh) OpenAI 78.3 — — 57.1%
89.9
2/4 $24.00/M 3.7
8 Muse Spark 1.3 (max) Meta 75.8 — — 58.8%
89.7
2/4 $5.50/M 16.3
9 GPT-5.6 Sol (high) OpenAI 77.2 — — 57.8%
89.7
2/4 $24.00/M 3.7
10 grok-4-fast-reasoning SpaceXAI — 1,420 83.2% —
89.7
2/4 $0.70/M 128.1
11 GPT-5.6 Sol (max) OpenAI 77.4 — — 57.1%
89.3
2/4 $24.00/M 3.7
12 GLM-5.3 (max) Z.ai 74.8 — — 59.0%
89.3
2/4 $4.90/M 18.2
13 Claude Opus 5 (Adaptive Reasoning, Max Effort) Anthropic 78.0 — — 56.4%
89.1
2/4 $30.00/M 3.0
14 GPT-5.6 Sol (medium) OpenAI 76.3 — — 57.4%
88.9
2/4 $24.00/M 3.7
15 Gemini 3.7 Flash (high) Google 76.1 — — 57.2%
88.6
2/4 $4.50/M 19.7
16 grok-4-1-fast-reasoning SpaceXAI — 1,411 82.2% —
88.6
2/4 — —
17 Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) Anthropic 77.1 — — 56.4%
88.5
2/4 $60.00/M 1.5
18 GPT-6 Astra (max) OpenAI 76.9 — — 56.5%
88.5
2/4 $60.00/M 1.5
19 Grok 4.6 (high) SpaceXAI 76.8 — — 56.5%
88.4
2/4 $8.00/M 11.1
20 Grok 4 SpaceXAI — 1,407 81.9% —
88.2
2/4 $18.00/M 4.9
21 Gemini 3.8 Flash (high) Google 76.3 — — 56.6%
88.2
2/4 $4.50/M 19.6
22 Gemini 3.7 Flash (medium) Google 71.5 — — 59.8%
87.9
2/4 $4.50/M 19.5
23 Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) Anthropic 77.0 — — 55.7%
87.9
2/4 $30.00/M 2.9
24 GPT-6 Astra (high) OpenAI 77.1 — — 55.4%
87.7
2/4 $60.00/M 1.5
25 Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) Anthropic 75.2 — — 56.7%
87.6
2/4 $60.00/M 1.5
26 Claude Opus 5 (Adaptive Reasoning, High Effort) Anthropic 76.5 — — 55.4%
87.3
2/4 $30.00/M 2.9
27 GPT-6 Astra (xhigh) OpenAI 75.9 — — 55.7%
87.2
2/4 $60.00/M 1.5
28 GPT-5.6 Terra (max) OpenAI 76.7 — — 55.0%
87.1
2/4 $14.00/M 6.2
29 Muse Spark 1.1 (xhigh) Meta 71.3 — — 58.8%
87.0
2/4 $5.50/M 15.8
30 GPT-5.5 (xhigh) OpenAI 74.9 — — 55.8%
86.7
2/4 $35.00/M 2.5
31 Grok 4.6 (medium) SpaceXAI 74.4 — — 55.9%
86.5
2/4 $8.00/M 10.8
32 GPT-6 Astra (medium) OpenAI 76.7 — — 54.2%
86.4
2/4 $60.00/M 1.4
33 Muse Spark 1.2 (xhigh) Meta 72.2 — — 57.4%
86.4
2/4 $5.50/M 15.7
34 MiniMax: MiniMax M2 MiniMax — 1,369 82.6% —
86.3
2/4 $1.28/M 67.7
35 GPT-6 Astra (low) OpenAI 75.7 — — 54.1%
85.8
2/4 $60.00/M 1.4
36 Gemini 3.8 Flash (medium) Google 74.1 — — 55.1%
85.6
2/4 $4.50/M 19.0
37 Google: Gemini 3.1 Pro Preview Google 68.8 — — 58.7%
85.3
2/4 $14.00/M 6.1
38 Claude Opus 4.8 (Adaptive Reasoning, Max Effort) Anthropic 74.3 — — 54.4%
85.2
2/4 $30.00/M 2.8
39 Gemini 3.8 Flash (low) Google 73.5 — — 55.0%
85.2
2/4 $4.50/M 18.9
40 Grok 4.6 (xhigh) SpaceXAI 75.9 — — 53.0%
84.9
2/4 $8.00/M 10.6
41 GPT-5.5 (high) OpenAI 71.6 — — 56.1%
84.9
2/4 $35.00/M 2.4
42 Grok 4.5 (high) SpaceXAI 72.4 — — 55.0%
84.5
2/4 $8.00/M 10.6
43 intellect-3 Prime Intellect — 1,383 77.7% —
84.5
2/4 — —
44 Qwen: Qwen3.8 Max (0902) Alibaba 76.2 — — 52.1%
84.4
2/4 $8.00/M 10.5
45 Gemini 3.7 Flash (low) Google 71.0 — — 55.7%
84.2
2/4 $4.50/M 18.7
46 GPT-5.6 Sol (low) OpenAI 69.7 — — 56.4%
84.0
2/4 $24.00/M 3.5
47 GPT-5.5 (medium) OpenAI 71.5 — — 54.5%
83.4
2/4 $35.00/M 2.4
48 Qwen: Qwen3.8 2.4T A95B Alibaba 71.9 — — 54.1%
83.4
2/4 $8.00/M 10.4
49 Claude Sonnet 5 (Adaptive Reasoning, Max Effort) Anthropic 71.5 — — 54.3%
83.3
2/4 $12.00/M 6.9
50 GPT-5 nano (high) OpenAI — 1,351 78.9% —
83.2
2/4 $0.45/M 184.8

Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) leads on raw performance, but GPT-5 nano (high) delivers 11450% more performance per dollar. That extra 13.7 points costs $4.35/M per point.

Access this data

JSON GET /api/v1 
[
  {
    "rank": 1,
    "model": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)",
    "provider": "Anthropic",
    "avg_score": 96.9,
    "cost": 60.0,
    "cost_unit": "million_tokens",
    "value": 1.6
  },
  {
    "rank": 2,
    "model": "Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)",
    "provider": "Anthropic",
    "avg_score": 94.5,
    "cost": 60.0,
    "cost_unit": "million_tokens",
    "value": 1.6
  },
  {
    "rank": 3,
    "model": "Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)",
    "provider": "Anthropic",
    "avg_score": 92.0,
    "cost": 60.0,
    "cost_unit": "million_tokens",
    "value": 1.5
  }
]
Curl 
$ curl -s https://rightrank.ai/api/v1/rankings/coding.json | jq '.[:3]'
CLI brew install rightrank 
$ rightrank top coding --limit 5
$ rightrank compare claude-opus-4.6 gpt-5.4-pro --category coding
$ rightrank value coding  # best performance per dollar
Claude Code Skill Skill 
# In Claude Code, use the RightRank skill:
/rightrank best model for coding
/rightrank compare top 5 by value
/rightrank cheapest model scoring above 80 on coding