Compare / Nemotron 3 Nano 30B A3B vs gpt-oss-20b

Nemotron 3 Nano 30B A3B vs gpt-oss-20b

Which is cheaper, which is more capable, and which does better at each kind of work, from 23 results on 3 sources that measured both models.

NVIDIANemotron 3 Nano 30B A3B Released 15 Dec 2025 · $0.060 in · $0.24 out
OpenAIgpt-oss-20b Released 5 Aug 2025 · $0.030 in · $0.14 out
Nemotron 3 Nano 30B A3B wins 87 too close to callgpt-oss-20b wins 8

Of the 23 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331).

Nemotron 3 Nano 30B A3B or gpt-oss-20b? The short answer

They score about the same, so pick gpt-oss-20b: it costs 45% less per token.

  • gpt-oss-20b is cheaper: $0.058 against $0.10 per million tokens, 45% less.
  • gpt-oss-20b scores higher on the Artificial Analysis Intelligence Index: 10.0 against 8.9.
  • Nemotron 3 Nano 30B A3B is newer: released 15 Dec 2025, against 5 Aug 2025 for gpt-oss-20b.
  • On third-party benchmarks, on the results both models have: Nemotron 3 Nano 30B A3B does better for reasoning and knowledge and writing; gpt-oss-20b for agents and tool use.

Head to head

8.9
Intelligence Index
10.0
$0.10
Price per million tokens, blendedlower is better
$0.058
238
Output speed, tokens a second
277
262,144
Context window, tokens
131,072

Every fact side by side

Nemotron 3 Nano 30B A3Bgpt-oss-20b
Price per million tokens$0.060 in · $0.24 out$0.030 in · $0.14 out
Blended (3 in : 1 out)$0.10$0.058
Cheapest host, blended–$0.022 (PrimeIntellect, FP4)
Intelligence Index8.910.0
Spring Prompt overall––
Output speed, tokens a second238277
Context window262,144 tokens131,072 tokens
Released15 Dec 2025 (newer)5 Aug 2025
WeightsOpenOpen
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 147 of 159 130 of 159
Reasoning and knowledgeHard questions across science, maths and general knowledge 131 of 187 132 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 153 of 169 155 of 169
Sticking to the factsSummarising without inventing things, and factual answers 111 of 139 not measured

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureNemotron 3 Nano 30B A3Bgpt-oss-20b
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 242 · 237 of 387 measured 1,314 1,318
Business, management and finance · TextArena rating, higher is better 223 · 213 of 380 measured 1,317 1,323
Creative writing · TextArena rating, higher is better 273 · 275 of 385 measured 1,242 1,240
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 191 · 176 of 249 measured 8.9thinking reasoning 10.0low reasoning
GPQA Diamond% of questions, higher is better 136 · 159 of 232 measured 75.7%thinking reasoning 68.8%high reasoning
Humanity's Last Exam% of questions, higher is better 151 · 153 of 250 measured 11.4%thinking reasoning 11.0%high reasoning
UGI Leaderboard reported by UGI Leaderboard · 3 results
Writing score · UGIscore out of 100, higher is better 173 · 182 of 230 measured 25.9 24.6low reasoning
Requested-length error · UGI% off the requested word count, lower is better 180 · 221 of 230 measured 35.0% 98.0%low reasoning
Style adherence · UGIscore from 0 to 1, higher is better 221 · 187 of 230 measured 0.28 0.31medium reasoning
Show all 23 results (14 more)
MeasureNemotron 3 Nano 30B A3Bgpt-oss-20b
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Expert prompts · TextArena rating, higher is better 208 · 220 of 338 measured 1,331 1,314
Instruction following · TextArena rating, higher is better 247 · 257 of 387 measured 1,291 1,281
Writing, literature and language · TextArena rating, higher is better 270 · 261 of 386 measured 1,267 1,275
Artificial Analysis reported by Artificial Analysis · 11 results
Terminal-Bench 2.1% of tasks, higher is better 108 · 97 of 131 measured 6.7%thinking reasoning 13.9%high reasoning
Terminal-Bench 4.0% of tasks, higher is better 72 · 72 of 120 measured 0.0%thinking reasoning 0.0%high reasoning
τ-bench banking% of tasks, higher is better 97 · 90 of 121 measured 6.0%thinking reasoning 7.0%high reasoning
Artificial Analysis Coding Indexindex score, higher is better 112 · 100 of 130 measured 14.4thinking reasoning 20.7high reasoning
IFBench% of instructions, higher is better 49 · 69 of 195 measured 71.1%thinking reasoning 65.1%high reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 180 · 184 of 241 measured 38.0%thinking reasoning 34.7%high reasoning
Output speedtokens per second, higher is better 11 · 6 of 78 measured 238thinking reasoning 277low reasoning
SciCode% of problems, higher is better 109 · 90 of 121 measured 30.6%thinking reasoning 38.9%high reasoning
Terminal-Bench Hard% of tasks, higher is better 116 · 127 of 190 measured 13.6%thinking reasoning 10.6%high reasoning
Time to first answer tokenseconds, lower is better 1 · 44 of 78 measured 0.36no reasoning 7.62low reasoning
τ²-bench telecom% of tasks, higher is better 115 · 96 of 193 measured 40.9%thinking reasoning 60.2%high reasoning

Quick answers

Is Nemotron 3 Nano 30B A3B better than gpt-oss-20b?

They score about the same, so pick gpt-oss-20b: it costs 45% less per token.

Which is cheaper, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

gpt-oss-20b is cheaper: $0.058 against $0.10 per million tokens, 45% less.

Which is faster, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

gpt-oss-20b writes about 277 tokens a second on its usual API, against 238 for Nemotron 3 Nano 30B A3B (Artificial Analysis).

Which is better for agents and tool use, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

gpt-oss-20b ranks 130 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 147 for Nemotron 3 Nano 30B A3B.

Which is better for reasoning and knowledge, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

Nemotron 3 Nano 30B A3B ranks 131 of 187 models on sale for hard questions across science, maths and general knowledge, against 132 for gpt-oss-20b.

Which is better for writing, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

Nemotron 3 Nano 30B A3B ranks 153 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 155 for gpt-oss-20b.

Which has the bigger context window, Nemotron 3 Nano 30B A3B or gpt-oss-20b?

Nemotron 3 Nano 30B A3B: 262,144 tokens, against 131,072 for gpt-oss-20b.

Run Nemotron 3 Nano 30B A3B and gpt-oss-20b on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.