DeepSeek-R1 vs gpt-oss-120b
Wins 2 of 6 areas
Facts · Long documents
Wins 4 of 6 areas
Coding · Agents · Reasoning · Following instructions
gpt-oss-120b is the stronger all-rounder.DeepSeek-R1 is better at facts.
Scores updated · 31 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- AgentsCarrying out multi-step tasks on its own02gpt-oss-120b2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02gpt-oss-120b2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked02gpt-oss-120b2 of 2 tests
- CodingWriting and fixing software12gpt-oss-120b2 of 3 tests
- FactsGetting facts right instead of making them up20DeepSeek-R12 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts10DeepSeek-R11 of 1 test
Long documents rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
gpt-oss-120b costs 88% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where gpt-oss-120b pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+30points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+17.4points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+11.1points ahead
Where DeepSeek-R1 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.7points ahead
- Reasons across sets of long documentsAA-LCR+5.7points ahead
- Code for real scientific research problemsSciCode+4.3points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Codinggpt-oss-120b
- Terminal-Bench Hardgpt-oss-120b by 17.46.123.5+17.4
- Terminal-Bench 2.1gpt-oss-120b by 7.119.126.2+7.1
- SciCodeDeepSeek-R1 by 4.338.334+4.3
Agentsgpt-oss-120b
- τ-Bench V3 · Bankinggpt-oss-120b by 6.46.412.8+6.4
- GDPValgpt-oss-120b by 5.605.6+5.6
- Terminal-Bench 4.0tie00tie
Reasoninggpt-oss-120b
- Humanity's Last Examgpt-oss-120b by 11.18.519.6+11.1
- GPQA Diamondgpt-oss-120b by 7.470.878.2+7.4
- CritPttie0.61.1tie
FactsDeepSeek-R1
- AA-Omniscience · AccuracyDeepSeek-R1 by 8.730.521.8+8.7
- Vectara HHEM hallucination ratelower is betterDeepSeek-R1 by 2.911.314.2+2.9
- AA-Omniscience · Non-hallucinationtie9.79.2tie
Following instructionsgpt-oss-120b
- IFBenchgpt-oss-120b by 303969+30
- Multi-Challengegpt-oss-120b by 4.640.745.3+4.6
Other results16 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- HMMT Feb. 2025gpt-oss-120b by 5541.796.7+55
- τ²-Bench Telecom (AA run)gpt-oss-120b by 54.411.465.8+54.4
- HarmBenchgpt-oss-120b by 52.147.9100+52.1
- AIR-Bench 2024gpt-oss-120b by 35.152.988+35.1
- AIME 2025gpt-oss-120b by 207090+20
- AA-OmniscienceDeepSeek-R1 by 17-32.2-49.3+17
- Aider-PolyglotDeepSeek-R1 by 11.553.341.8+11.5
- Artificial Analysis Coding Indexgpt-oss-120b by 5.824.630.4+5.8
- vectara_avg_summary_lengthgpt-oss-120b by 41.7 rating points93.5135.2+41.7 rating
- vectara_answer_rategpt-oss-120b by 2.99799.9+2.9
- vectara_factual_consistencyDeepSeek-R1 by 2.988.785.8+2.9
- anthropic_red_teamgpt-oss-120b by 2.397.299.5+2.3
- simple_safety_testsgpt-oss-120b by 298100+2
- bbqgpt-oss-120b by 1.996.698.5+1.9
- XSTestDeepSeek-R1 by 1.794.492.7+1.7
- AA Intelligencetie11.411.6tie
Questions people ask
Which is better, DeepSeek-R1 or gpt-oss-120b?
gpt-oss-120b wins four of the six areas where both have results: coding, agents, reasoning and following instructions. DeepSeek-R1 wins facts and long documents.
Which is better for coding?
gpt-oss-120b. It wins 2 of the 3 coding tests both models report; DeepSeek-R1 wins 1.
Which is cheaper?
DeepSeek-R1 costs $2.00 per million input tokens and $4.00 per million output tokens; gpt-oss-120b costs $0.15 and $0.59. That makes gpt-oss-120b about 88% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.