Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 5.5 vs GPT-5.6 Terra
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 5.5 wins on 16/17 benchmarks
Claude Sonnet 5.5 wins 16 of 17 shared benchmarks. Leads in speed · agentic · knowledge.
Category leads
speed·Claude Sonnet 5.5agentic·Claude Sonnet 5.5knowledge·Claude Sonnet 5.5coding·Claude Sonnet 5.5math·Claude Sonnet 5.5general·Claude Sonnet 5.5
Hype vs Reality
Attention vs performance
Claude Sonnet 5.5
#34 by perf·#12 by attention
GPT-5.6 Terra
#32 by perf·#10 by attention
Best value
Claude Sonnet 5.5
1.2x better value than GPT-5.6 Terra
Claude Sonnet 5.5
11.0 pts/$
$6.00/M
GPT-5.6 Terra
9.5 pts/$
$7.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
17 benchmarks · 2 models
Claude Sonnet 5.5GPT-5.6 Terra
Artificial Analysis · CritPt
Claude Sonnet 5.5 leads by +1.4
Claude Sonnet 5.5
31.4
GPT-5.6 Terra
30.0
Artificial Analysis · GDPval
Claude Sonnet 5.5 leads by +19.3
Claude Sonnet 5.5
66.9
GPT-5.6 Terra
47.6
Artificial Analysis · Humanity's Last Exam
Claude Sonnet 5.5 leads by +12.1
Claude Sonnet 5.5
55.0
GPT-5.6 Terra
42.9
Artificial Analysis · Long Context Reasoning
GPT-5.6 Terra leads by +0.3
Claude Sonnet 5.5
82.7
GPT-5.6 Terra
83.0
Artificial Analysis · Quality Index
Claude Sonnet 5.5 leads by +13.9
Claude Sonnet 5.5
56.0
GPT-5.6 Terra
42.1
Artificial Analysis · SciCode
Claude Sonnet 5.5 leads by +6.0
Claude Sonnet 5.5
61.0
GPT-5.6 Terra
55.0
APEX-Agents
Claude Sonnet 5.5 leads by +17.3
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Sonnet 5.5
75.5
GPT-5.6 Terra
58.2
Chess Puzzles
Claude Sonnet 5.5 leads by +12.6
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Sonnet 5.5
56.9
GPT-5.6 Terra
44.2
Frontiercode
Claude Sonnet 5.5 leads by +10.8
Claude Sonnet 5.5
52.1
GPT-5.6 Terra
41.3
FrontierMath-Tier-4-v2-Private
Claude Sonnet 5.5 leads by +9.8
Claude Sonnet 5.5
80.5
GPT-5.6 Terra
70.7
FrontierMath-Tiers-1-3-v2-Private
Claude Sonnet 5.5 leads by +2.8
Claude Sonnet 5.5
88.8
GPT-5.6 Terra
86.0
Furniture Assembly
Claude Sonnet 5.5 leads by +29.8
Claude Sonnet 5.5
64.3
GPT-5.6 Terra
34.5
GPQA diamond
Claude Sonnet 5.5 leads by +3.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 5.5
94.1
GPT-5.6 Terra
91.1
Mystery Game Puzzles
Claude Sonnet 5.5 leads by +33.1
Claude Sonnet 5.5
61.5
GPT-5.6 Terra
28.4
OTIS Mock AIME 2024-2025
Claude Sonnet 5.5 leads by +0.3
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 5.5
100.0
GPT-5.6 Terra
99.7
Proofbench
Claude Sonnet 5.5 leads by +26.0
Claude Sonnet 5.5
100.0
GPT-5.6 Terra
74.0
SimpleQA Verified
Claude Sonnet 5.5 leads by +3.3
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Sonnet 5.5
46.5
GPT-5.6 Terra
43.2
Full benchmark table
| Benchmark | Claude Sonnet 5.5 | GPT-5.6 Terra |
|---|---|---|
Artificial Analysis · CritPt | 31.4 | 30.0 |
Artificial Analysis · GDPval | 66.9 | 47.6 |
Artificial Analysis · Humanity's Last Exam | 55.0 | 42.9 |
Artificial Analysis · Long Context Reasoning | 82.7 | 83.0 |
Artificial Analysis · Quality Index | 56.0 | 42.1 |
Artificial Analysis · SciCode | 61.0 | 55.0 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 75.5 | 58.2 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 56.9 | 44.2 |
Frontiercode | 52.1 | 41.3 |
FrontierMath-Tier-4-v2-Private | 80.5 | 70.7 |
FrontierMath-Tiers-1-3-v2-Private | 88.8 | 86.0 |
Furniture Assembly | 64.3 | 34.5 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 94.1 | 91.1 |
Mystery Game Puzzles | 61.5 | 28.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 100.0 | 99.7 |
Proofbench | 100.0 | 74.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 46.5 | 43.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $10.00 | 1.0M tokens (~500 books) | $40.00 | |
| $2.00 | $12.00 | 1.1M tokens (~525 books) | $45.00 |
People also compared
GPT-5.6 Terra vs phi-3-small 7.4BClaude Sonnet 5.5 vs phi-3-small 7.4BClaude Sonnet 5.5 vs GPT-5.6 SolGPT-5.6 Terra vs Qwen3 30B A3B Thinking 2507Claude Sonnet 5.5 vs gpt-oss-20b (free)Claude Sonnet 5.5 vs Qwen3 30B A3B Thinking 2507Claude Sonnet 5.5 vs GPT-4 Turbo (older v1106)GPT-5.6 Terra vs Grok 3 Mini Beta