Compare / Nova 2 Lite vs GPT-4.1 Mini
Nova 2 Lite vs GPT-4.1 Mini
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.
Nova 2 Lite or GPT-4.1 Mini? The short answer
- GPT-4.1 Mini is cheaper: $0.70 against $0.85 per million tokens, 18% less.
- Nova 2 Lite scores higher on the Artificial Analysis Intelligence Index: 13.4 against 10.2.
- On our business benchmarks, Nova 2 Lite ranks higher for reasoning and knowledge and sticking to the facts.
- On our business benchmarks, GPT-4.1 Mini ranks higher for agents and tool use and writing.
At a glance
| Nova 2 Lite | GPT-4.1 Mini | |
|---|---|---|
| Price per million tokens | $0.30 in · $2.50 out | $0.40 in · $1.60 out |
| Blended (3 in : 1 out) | $0.85 | $0.70 |
| Cheapest host, blended | $0.85 (Amazon Bedrock) | $0.70 (Azure) |
| Intelligence Index | 13.4 | 10.2 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | 210 | – |
| Context window | 1,000,000 tokens | 1,047,576 tokens |
| Weights | Closed | Closed |
| 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 | Nova 2 Lite | GPT-4.1 Mini |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 130 of 161 | 105 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 112 of 195 | 147 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 168 of 187 | 122 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | 76 of 153 | 144 of 153 |
Every shared result
| Benchmark | Metric | Nova 2 Lite | GPT-4.1 Mini |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,351 | 1,379 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,269 | 1,349 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,354 | 1,384 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,327 | 1,373 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,335 | 1,383 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,298 | 1,361 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 23.0 | 20.2 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 13.4 | 10.2 |
| Artificial Analysisreported | GPQA Diamond% of questions | 81.1% | 66.4% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 11.6% | 5.0% |
| Artificial Analysisreported | IFBench% of instructions | 70.7% | 38.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 60.3% | 44.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 16.1% | 10.1% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 17.4% | 7.6% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 75.7% | 52.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 82.1% | 81.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 2.4% | 26.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 2.1% | 34.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 27.1% | 50.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 75.0% | 87.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 87.0% | 84.6% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 80.8% | 74.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 5.0% | 57.0% |
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 Nova 2 Lite better than GPT-4.1 Mini?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Nova 2 Lite ranks higher for reasoning and knowledge and sticking to the facts; GPT-4.1 Mini ranks higher for agents and tool use and writing.
Which is cheaper, Nova 2 Lite or GPT-4.1 Mini?
GPT-4.1 Mini costs $0.70 per million tokens (three input to one output), against $0.85 for Nova 2 Lite.
Run Nova 2 Lite and GPT-4.1 Mini on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.