Compare · ModelsLive · 2 picked · head to head
Claude Sonnet 4.5 vs gpt-oss-20b
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Sonnet 4.5 wins on 7/7 benchmarks
Claude Sonnet 4.5 wins 7 of 7 shared benchmarks. Leads in knowledge · general · math.
Category leads
knowledge·Claude Sonnet 4.5general·Claude Sonnet 4.5math·Claude Sonnet 4.5coding·Claude Sonnet 4.5
Hype vs Reality
Attention vs performance
Claude Sonnet 4.5
#207 by perf·#10 by attention
gpt-oss-20b
#97 by perf·no signal
Best value
gpt-oss-20b
234.1x better value than Claude Sonnet 4.5
Claude Sonnet 4.5
4.2 pts/$
$9.00/M
gpt-oss-20b
983.3 pts/$
$0.05/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
7 benchmarks · 2 models
Claude Sonnet 4.5gpt-oss-20b
Chess Puzzles
Claude Sonnet 4.5 leads by +7.4
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Sonnet 4.5
7.4
gpt-oss-20b
0.0
Dtbench
Claude Sonnet 4.5 leads by +25.3
Claude Sonnet 4.5
72.0
gpt-oss-20b
46.7
GPQA diamond
Claude Sonnet 4.5 leads by +28.7
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Sonnet 4.5
76.4
gpt-oss-20b
47.7
Lmca
Claude Sonnet 4.5 leads by +28.5
Claude Sonnet 4.5
45.6
gpt-oss-20b
17.1
OTIS Mock AIME 2024-2025
Claude Sonnet 4.5 leads by +12.5
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 4.5
77.8
gpt-oss-20b
65.2
Terminal Bench
Claude Sonnet 4.5 leads by +43.1
Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency. Claude Opus 4.7 scores 69.4%, demonstrating significant agentic terminal competence.
Claude Sonnet 4.5
46.5
gpt-oss-20b
3.4
WeirdML
Claude Sonnet 4.5 leads by +6.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Sonnet 4.5
47.7
gpt-oss-20b
40.9
Full benchmark table
| Benchmark | Claude Sonnet 4.5 | gpt-oss-20b |
|---|---|---|
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 7.4 | 0.0 |
Dtbench | 72.0 | 46.7 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 76.4 | 47.7 |
Lmca | 45.6 | 17.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 77.8 | 65.2 |
Terminal Bench Terminal-Bench 2.0 · evaluates AI agents on real terminal-based coding tasks · writing scripts, debugging, running tests, and managing projects entirely through command-line interaction. Tests both code quality and terminal fluency. Claude Opus 4.7 scores 69.4%, demonstrating significant agentic terminal competence. | 46.5 | 3.4 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 47.7 | 40.9 |
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 | |
| $0.02 | $0.09 | 131K tokens (~66 books) | $0.36 |