Compare · ModelsLive · 2 picked · head to head

Claude Sonnet 4.5 vs gpt-oss-120b

Side by side · benchmarks, pricing, and signals you can act on.

Winner summary

Claude Sonnet 4.5 wins 9 of 12 shared benchmarks. Leads in knowledge · general · reasoning.

Category leads
knowledge·Claude Sonnet 4.5general·Claude Sonnet 4.5math·gpt-oss-120breasoning·Claude Sonnet 4.5coding·Claude Sonnet 4.5

Head to head · from the data · data as of 2026-11-18

  • Benchmarks · Claude Sonnet 4.5 leads on 9 of 12 shared benchmarks (Dtbench, GPQA diamond, Lmca, Metr Time Horizons, 5 more), gpt-oss-120b on 3 (Chess Puzzles, OTIS Mock AIME 2024-2025, WeirdML)
  • Price · Claude Sonnet 4.5 input $3.00/M vs gpt-oss-120b $0.15/M, gpt-oss-120b is 95% cheaper (OpenRouter list price, as of 2026-11-18)
  • Cheapest provider · Claude Sonnet 4.5 at Amazon Bedrock and 4 other providers $3.00/M input (of 7 offers) as of 2026-11-18 · gpt-oss-120b at AkashML and 4 other providers $0.03/M input (of 24 offers) as of 2026-11-18
  • Context · Claude Sonnet 4.5 1.0M tokens vs gpt-oss-120b 131K tokens
  • Released · Claude Sonnet 4.5 Sep 29, 2025 vs gpt-oss-120b Aug 5, 2025 · 1 month apart
  • Open weights · gpt-oss-120b has open weights (on Hugging Face as openai/gpt-oss-120b)
  • Price history · Claude Sonnet 4.5 OpenRouter input list price had no move of 5% or more in the 90 days to 2026-11-18

Sources · scores from public leaderboards, list prices and provider offers from OpenRouter, price moves from the BenchGecko changelog (5% or more, held in the next daily check), Gecko tests run by BenchGecko.

Hype vs Reality
Claude Sonnet 4.5
#217 by perf·#12 by attention
QUIET
gpt-oss-120b
#142 by perf·no signal
QUIET
Best value
30.7x better value than Claude Sonnet 4.5
Claude Sonnet 4.5
4.3 pts/$
$9.00/M
gpt-oss-120b
131.5 pts/$
$0.38/M
Vendor risk
Anthropic logo
Anthropic
$1.40T·Tier 1
Medium risk
OpenAI logo
OpenAI
$840.0B·Tier 1
Medium risk
Head to head
Claude Sonnet 4.5gpt-oss-120b
Chess Puzzles
gpt-oss-120b leads by +10.5
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Sonnet 4.5
5.3
gpt-oss-120b
15.8
Dtbench
Claude Sonnet 4.5 leads by +11.5
Claude Sonnet 4.5
72.0
gpt-oss-120b
60.5
GPQA diamond
Claude Sonnet 4.5 leads by +11.9
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
79.6
gpt-oss-120b
67.7
Lmca
Claude Sonnet 4.5 leads by +19.6
Claude Sonnet 4.5
45.6
gpt-oss-120b
26.1
Metr Time Horizons
Claude Sonnet 4.5 leads by +10.7
Claude Sonnet 4.5
67.4
gpt-oss-120b
56.6
Mystery Game Puzzles
Claude Sonnet 4.5 leads by +8.6
Claude Sonnet 4.5
8.6
gpt-oss-120b
0.0
OTIS Mock AIME 2024-2025
gpt-oss-120b leads by +11.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Sonnet 4.5
77.8
gpt-oss-120b
88.9
SimpleBench
Claude Sonnet 4.5 leads by +38.6
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Sonnet 4.5
45.2
gpt-oss-120b
6.5
SimpleQA Verified
Claude Sonnet 4.5 leads by +16.8
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Sonnet 4.5
30.7
gpt-oss-120b
13.9
SWE-Bench Verified (Bash Only)
Claude Sonnet 4.5 leads by +44.6
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 4.5
70.6
gpt-oss-120b
26.0
Terminal Bench
Claude Sonnet 4.5 leads by +27.8
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 Sonnet 4.5
46.5
gpt-oss-120b
18.7
WeirdML
gpt-oss-120b leads by +0.5
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-120b
48.2
Full benchmark table
BenchmarkClaude Sonnet 4.5gpt-oss-120b
Chess Puzzles
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
5.315.8
Dtbench
72.060.5
GPQA diamond
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
79.667.7
Lmca
45.626.1
Metr Time Horizons
67.456.6
Mystery Game Puzzles
8.60.0
OTIS Mock AIME 2024-2025
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
77.888.9
SimpleBench
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
45.26.5
SimpleQA Verified
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
30.713.9
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.
70.626.0
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.
46.518.7
WeirdML
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
47.748.2
Pricing · per 1M tokens · projected $/mo at 10M tokens
ModelInputOutputContextProjected $/mo
Anthropic logoClaude Sonnet 4.5$3.00$15.001.0M tokens (~500 books)$60.00
OpenAI logogpt-oss-120b$0.15$0.60131K tokens (~66 books)$2.62
Claude Sonnet 4.5 leads on 9 of 12 shared benchmarks (Dtbench, GPQA diamond, Lmca, Metr Time Horizons, 5 more), gpt-oss-120b on 3 (Chess Puzzles, OTIS Mock AIME 2024-2025, WeirdML)