Compare / GPT-5 nano vs gpt-oss-20b
GPT-5 nano vs gpt-oss-20b
Which is cheaper, which is more capable, and which does better at each kind of work, from 13 results on 2 sources that measured both models.
Of the 13 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).
GPT-5 nano or gpt-oss-20b? The short answer
Pick GPT-5 nano if quality matters most: it scores higher (Intelligence Index 13.0 against 10.0). Pick gpt-oss-20b if cost matters more: it costs 2.4 times less per token.
- gpt-oss-20b is cheaper: $0.058 against $0.14 per million tokens, 2.4 times less.
- GPT-5 nano scores higher on the Artificial Analysis Intelligence Index: 13.0 against 10.0.
- GPT-5 nano is newer: released 7 Aug 2025, against 5 Aug 2025 for gpt-oss-20b.
- On third-party benchmarks, on the results both models have: GPT-5 nano does better for writing; gpt-oss-20b for reasoning and knowledge.
Head to head
Every fact side by side
| GPT-5 nano | gpt-oss-20b | |
|---|---|---|
| Price per million tokens | $0.050 in · $0.40 out | $0.030 in · $0.14 out |
| Blended (3 in : 1 out) | $0.14 | $0.058 |
| Cheapest host, blended | $0.14 (Azure) | $0.022 (PrimeIntellect, FP4) |
| Intelligence Index | 13.0 | 10.0 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | 277 |
| Context window | 400,000 tokens | 131,072 tokens |
| Released | 7 Aug 2025 (newer) | 5 Aug 2025 |
| Weights | Closed | Open |
| Developer based in | the US | the 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.
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.
| Measure | GPT-5 nano | gpt-oss-20b |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Overall · TextArena rating, higher is better 207 · 237 of 387 measured | 1,338high reasoning | 1,318 |
| Business, management and finance · TextArena rating, higher is better 204 · 213 of 380 measured | 1,339high reasoning | 1,323 |
| Creative writing · TextArena rating, higher is better 268 · 275 of 385 measured | 1,247high reasoning | 1,240 |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 149 · 176 of 249 measured | 13.0high reasoning | 10.0low reasoning |
| GPQA Diamond% of questions, higher is better 163 · 159 of 232 measured | 67.6%high reasoning | 68.8%high reasoning |
| Humanity's Last Exam% of questions, higher is better 168 · 153 of 250 measured | 9.5%high reasoning | 11.0%high reasoning |
Show all 13 results (7 more)
| Measure | GPT-5 nano | gpt-oss-20b |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Expert prompts · TextArena rating, higher is better 191 · 220 of 338 measured | 1,354high reasoning | 1,314 |
| Instruction following · TextArena rating, higher is better 200 · 257 of 387 measured | 1,327high reasoning | 1,281 |
| Writing, literature and language · TextArena rating, higher is better 247 · 261 of 386 measured | 1,289high reasoning | 1,275 |
| Artificial Analysis reported by Artificial Analysis · 4 results | ||
| IFBench% of instructions, higher is better 64 · 69 of 195 measured | 67.6%high reasoning | 65.1%high reasoning |
| Long-context reasoning (AA-LCR)% of questions, higher is better 170 · 184 of 241 measured | 45.0%high reasoning | 34.7%high reasoning |
| Terminal-Bench Hard% of tasks, higher is better 100 · 127 of 190 measured | 17.4%medium reasoning | 10.6%high reasoning |
| τ²-bench telecom% of tasks, higher is better 119 · 96 of 193 measured | 36.6%high reasoning | 60.2%high reasoning |
Quick answers
Is GPT-5 nano better than gpt-oss-20b?
Pick GPT-5 nano if quality matters most: it scores higher (Intelligence Index 13.0 against 10.0). Pick gpt-oss-20b if cost matters more: it costs 2.4 times less per token.
Which is cheaper, GPT-5 nano or gpt-oss-20b?
gpt-oss-20b is cheaper: $0.058 against $0.14 per million tokens, 2.4 times less.
Which is better for agents and tool use, GPT-5 nano or gpt-oss-20b?
GPT-5 nano ranks 57 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 130 for gpt-oss-20b.
Which is better for reasoning and knowledge, GPT-5 nano or gpt-oss-20b?
GPT-5 nano ranks 129 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, GPT-5 nano or gpt-oss-20b?
GPT-5 nano ranks 133 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, GPT-5 nano or gpt-oss-20b?
GPT-5 nano: 400,000 tokens, against 131,072 for gpt-oss-20b.
Run GPT-5 nano 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.