Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 4 vs gpt-oss-120b
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 4 wins on 6/8 benchmarks
Claude Sonnet 4 wins 6 of 8 shared benchmarks. Leads in coding · agentic · knowledge.
Category leads
coding·Claude Sonnet 4agentic·Claude Sonnet 4knowledge·Claude Sonnet 4math·gpt-oss-120breasoning·Claude Sonnet 4
Hype vs Reality
Attention vs performance
Claude Sonnet 4
#151 by perf·no signal
gpt-oss-120b
#137 by perf·no signal
Best value
gpt-oss-120b
90.3x better value than Claude Sonnet 4
Claude Sonnet 4
4.8 pts/$
$9.00/M
gpt-oss-120b
434.3 pts/$
$0.11/M
Vendor risk
Who is behind the model
Anthropic
$380.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
8 benchmarks · 2 models
Claude Sonnet 4gpt-oss-120b
Aider polyglot
Claude Sonnet 4 leads by +19.5
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
Claude Sonnet 4
61.3
gpt-oss-120b
41.8
APEX-Agents
Claude Sonnet 4 leads by +4.6
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Sonnet 4
9.3
gpt-oss-120b
4.7
Fiction.LiveBench
Claude Sonnet 4 leads by +2.5
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
Claude Sonnet 4
46.9
gpt-oss-120b
44.4
GPQA diamond
Claude Sonnet 4 leads by +4.6
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 4
72.3
gpt-oss-120b
67.7
OTIS Mock AIME 2024-2025
gpt-oss-120b leads by +17.8
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 4
71.1
gpt-oss-120b
88.9
SimpleBench
Claude Sonnet 4 leads by +28.1
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Sonnet 4
34.6
gpt-oss-120b
6.5
SWE-Bench Verified (Bash Only)
Claude Sonnet 4 leads by +38.9
SWE-Bench Verified (Bash Only) · a curated subset of SWE-bench where models fix real Python repository bugs using only bash commands, no agent frameworks.
Claude Sonnet 4
64.9
gpt-oss-120b
26.0
WeirdML
gpt-oss-120b leads by +2.1
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Sonnet 4
46.1
gpt-oss-120b
48.2
Full benchmark table
| Benchmark | Claude Sonnet 4 | gpt-oss-120b |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 61.3 | 41.8 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 9.3 | 4.7 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 46.9 | 44.4 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 72.3 | 67.7 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 71.1 | 88.9 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 34.6 | 6.5 |
SWE-Bench Verified (Bash Only) SWE-Bench Verified (Bash Only) · a curated subset of SWE-bench where models fix real Python repository bugs using only bash commands, no agent frameworks. | 64.9 | 26.0 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 46.1 | 48.2 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $3.00 | $15.00 | 1.0M tokens (~500 books) | $60.00 | |
| $0.04 | $0.18 | 131K tokens (~66 books) | $0.72 |