GPT-5 nano vs GPT-OSS 20B

Change models
IndexAll categories, out of 100.
41.9 ±6.2 best
41.7 ±4.6
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
75%
90% best
SpeedOutput tokens per second.
81/s
310/s best
EloLMArena rating and rank.
1320 best
1288
ReleasedMonth of release.
2025-08
2025-08
Input / 1MUS dollars per 1M input tokens.
$0.05
$0.03 best
Output / 1MUS dollars per 1M output tokens.
$0.40
$0.13 best
Batch output / 1MBatch price per 1M output tokens.
$0.20
—
1K chats2K tokens in, 500 out, each.
$0.30
$0.125 best
1K documents20K tokens in, 1K out, each.
$1.40
$0.73 best
1K agent tasks200K tokens in, 20K out, each.
$18.00
$8.60 best
ContextMaximum tokens in one request.
400K best
131K
AgenticMulti-step tasks with tools.
—
39.8
CodingCode writing and repair.
37.2
48.9 best
ReasoningLogic problems and puzzles.
42.6 best
35.3
MultimodalTasks with images and text.
47.5
—
KnowledgeFacts and expert knowledge.
37.4 best
36.0
MultilingualTasks in many languages.
44.5 best
37.0
InstructionTasks with strict prompt rules.
—
49.9
MathMath problems.
45.8
45.9 best

Benchmarks

20 shared
Only GPT-5 nano: 20Only GPT-OSS 20B: 17
GPT-5 nano
GPT-OSS 20B
69.4% best114 / 217
46.0%161 / 217
EuroEval PolishMultilingual
52.8% best56 / 209
50.2%66 / 209
81.1% best85 / 200
50.8%122 / 200
EuroEval SwedishMultilingual
62.3% best29 / 195
55.5%50 / 195
EuroEval GermanMultilingual
47.0% best51 / 191
45.8%57 / 191
49.0%73 / 182
50.6% best66 / 182
EuroEval FrenchMultilingual
53.9% best70 / 178
52.9%76 / 178
EuroEval SpanishMultilingual
49.0% best48 / 176
45.8%59 / 176
EuroEval DutchMultilingual
53.7% best49 / 175
49.9%70 / 175
EuroEval ItalianMultilingual
53.8% best49 / 174
51.0%58 / 174
50.1%147 / 149
70.8% best122 / 149
TaxEval v2Finance
67.4% best107 / 143
63.7%119 / 143
70.2%88 / 142
80.4% best69 / 142
63.4%105 / 135
68.9% best97 / 135
MMLU ProKnowledge
76.1% best107 / 135
71.6%115 / 135
AIMEMath
81.2%45 / 95
86.0% best33 / 95
MedQAHealthcare
93.3% best23 / 95
82.9%65 / 95
MGSMMultilingual
89.3% best47 / 74
89.0%50 / 74
93.8%15 / 59
94.2% best10 / 59
52.6% best42 / 53
43.8%53 / 53