Compare · ModelsLive · 2 picked · head to head
gpt-oss-120b vs Grok 4
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Grok 4 wins on 12/17 benchmarks
Grok 4 wins 12 of 17 shared benchmarks. Leads in coding · agentic · knowledge.
Category leads
coding·Grok 4agentic·Grok 4knowledge·Grok 4language·Grok 4math·gpt-oss-120breasoning·gpt-oss-120bgeneral·Grok 4
Hype vs Reality
Attention vs performance
gpt-oss-120b
#127 by perf·no signal
Grok 4
#87 by perf·#15 by attention
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
xAI
$250.0B·Tier 1
Head to head
17 benchmarks · 2 models
gpt-oss-120bGrok 4
Aider polyglot
Grok 4 leads by +37.8
Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework.
gpt-oss-120b
41.8
Grok 4
79.6
APEX-Agents
Grok 4 leads by +10.8
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
gpt-oss-120b
4.4
Grok 4
15.2
Chess Puzzles
Grok 4 leads by +8.4
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
gpt-oss-120b
15.8
Grok 4
24.2
Fiction.LiveBench
Grok 4 leads by +50.0
Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination.
gpt-oss-120b
44.4
Grok 4
94.4
GPQA diamond
Grok 4 leads by +15.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
gpt-oss-120b
67.7
Grok 4
82.7
HELM · GPQA
Grok 4 leads by +4.2
gpt-oss-120b
68.4
Grok 4
72.6
HELM · IFEval
Grok 4 leads by +11.3
gpt-oss-120b
83.6
Grok 4
94.9
HELM · MMLU-Pro
Grok 4 leads by +5.6
gpt-oss-120b
79.5
Grok 4
85.1
HELM · Omni-MATH
gpt-oss-120b leads by +8.5
gpt-oss-120b
68.8
Grok 4
60.3
HELM · WildBench
gpt-oss-120b leads by +4.8
gpt-oss-120b
84.5
Grok 4
79.7
Lech Mazur Writing
gpt-oss-120b leads by +0.2
Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication.
gpt-oss-120b
77.1
Grok 4
76.9
Metr Time Horizons
Grok 4 leads by +9.9
gpt-oss-120b
56.6
Grok 4
66.6
OTIS Mock AIME 2024-2025
gpt-oss-120b leads by +4.9
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
gpt-oss-120b
88.9
Grok 4
84.0
SimpleBench
Grok 4 leads by +46.1
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
gpt-oss-120b
6.5
Grok 4
52.6
SimpleQA Verified
Grok 4 leads by +34.0
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
gpt-oss-120b
13.9
Grok 4
47.9
Terminal Bench
Grok 4 leads by +8.5
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.
gpt-oss-120b
18.7
Grok 4
27.2
WeirdML
gpt-oss-120b leads by +2.4
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
gpt-oss-120b
48.2
Grok 4
45.7
Full benchmark table
| Benchmark | gpt-oss-120b | Grok 4 |
|---|---|---|
Aider polyglot Aider Polyglot · measures how well AI models can edit code across multiple programming languages using the Aider coding assistant framework. | 41.8 | 79.6 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 4.4 | 15.2 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 15.8 | 24.2 |
Fiction.LiveBench Fiction.LiveBench · a continuously updated benchmark using recently published fiction to test reading comprehension and reasoning, preventing data contamination. | 44.4 | 94.4 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 67.7 | 82.7 |
HELM · GPQA | 68.4 | 72.6 |
HELM · IFEval | 83.6 | 94.9 |
HELM · MMLU-Pro | 79.5 | 85.1 |
HELM · Omni-MATH | 68.8 | 60.3 |
HELM · WildBench | 84.5 | 79.7 |
Lech Mazur Writing Lech Mazur Writing · evaluates creative writing ability, assessing prose quality, narrative coherence, and stylistic sophistication. | 77.1 | 76.9 |
Metr Time Horizons | 56.6 | 66.6 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 88.9 | 84.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 6.5 | 52.6 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 13.9 | 47.9 |
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. | 18.7 | 27.2 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 48.2 | 45.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.04 | $0.17 | 131K tokens (~66 books) | $0.70 | |
| — | — | — | — |
People also compared