Compare / Llama 3.3 70B Instruct vs gpt-oss-20b
Llama 3.3 70B Instruct vs gpt-oss-20b
Which is cheaper, which is more capable, and which does better at each kind of work, from 20 results on 3 sources that measured both models.
Llama 3.3 70B Instruct or gpt-oss-20b? The short answer
- gpt-oss-20b is cheaper: $0.060 against $0.64 per million tokens, 11 times less.
- gpt-oss-20b scores higher on the Artificial Analysis Intelligence Index: 10.0 against 7.7.
- gpt-oss-20b writes faster: 249 tokens a second against 80.
- On our business benchmarks, Llama 3.3 70B Instruct ranks higher for writing.
- On our business benchmarks, gpt-oss-20b ranks higher for agents and tool use and reasoning and knowledge.
At a glance
| Llama 3.3 70B Instruct | gpt-oss-20b | |
|---|---|---|
| Price per million tokens | $0.59 in · $0.79 out | $0.030 in · $0.15 out |
| Blended (3 in : 1 out) | $0.64 | $0.060 |
| Cheapest host, blended | $0.16 (DeepInfra, FP8) | $0.036 (Darkbloom, FP8) |
| Intelligence Index | 7.7 | 10.0 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | 80 | 249 |
| Context window | 131,072 tokens | 131,072 tokens |
| Weights | Open | 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 models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Llama 3.3 70B Instruct | gpt-oss-20b |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 127 of 161 | 124 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 174 of 195 | 134 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 165 of 187 | 170 of 187 |
Every shared result
| Benchmark | Metric | Llama 3.3 70B Instruct | gpt-oss-20b |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,304 | 1,323 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,285 | 1,240 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,298 | 1,314 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,293 | 1,281 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,318 | 1,318 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,292 | 1,275 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 11.9 | 20.7 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 7.70 | 10.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 49.8% | 68.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 3.6% | 11.0% |
| Artificial Analysisreported | IFBench% of instructions | 47.1% | 65.1% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 15.7% | 34.7% |
| Artificial Analysisreported | Output speedtokens per second | 80.5 | 249 |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 4.9% | 13.9% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 3.0% | 10.6% |
| Artificial Analysisreported | Time to first answer tokenseconds | 0.64 | 8.49 |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 26.6% | 60.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 10.0% | 98.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.37 | 0.31 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 26.2 | 24.6 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Llama 3.3 70B Instruct better than gpt-oss-20b?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Llama 3.3 70B Instruct ranks higher for writing; gpt-oss-20b ranks higher for agents and tool use and reasoning and knowledge.
Which is cheaper, Llama 3.3 70B Instruct or gpt-oss-20b?
gpt-oss-20b costs $0.060 per million tokens (three input to one output), against $0.64 for Llama 3.3 70B Instruct.
Which is faster, Llama 3.3 70B Instruct or gpt-oss-20b?
gpt-oss-20b writes about 249 tokens a second on its usual API, against 80 for Llama 3.3 70B Instruct (Artificial Analysis).
Run Llama 3.3 70B Instruct and gpt-oss-20b on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.