Compare / Llama 3.1 70B Instruct vs Phi 4
Llama 3.1 70B Instruct vs Phi 4
Which is cheaper, which is more capable, and which does better at each kind of work, from 15 results on 3 sources that measured both models.
Llama 3.1 70B Instruct or Phi 4? The short answer
- Phi 4 is cheaper: $0.088 against $0.40 per million tokens, 4.6 times less.
- They score about the same on the Artificial Analysis Intelligence Index (6.6 and 5.9).
- On our business benchmarks, Llama 3.1 70B Instruct ranks higher for writing.
- On our business benchmarks, Phi 4 ranks higher for reasoning and knowledge.
At a glance
| Llama 3.1 70B Instruct | Phi 4 | |
|---|---|---|
| Price per million tokens | $0.40 in · $0.40 out | $0.070 in · $0.14 out |
| Blended (3 in : 1 out) | $0.40 | $0.088 |
| Cheapest host, blended | $0.40 (DeepInfra, FP8) | – |
| Intelligence Index | 6.6 | 5.9 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | 42 |
| Context window | 131,072 tokens | 16,384 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 3.1 70B Instruct | Phi 4 |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | not measured | 122 of 161 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 179 of 195 | 178 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 153 of 187 | 175 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | not measured | 3 of 153 |
Every shared result
| Benchmark | Metric | Llama 3.1 70B Instruct | Phi 4 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,287 | 1,256 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,257 | 1,210 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,275 | 1,268 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,273 | 1,245 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,293 | 1,256 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,264 | 1,229 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 6.60 | 5.90 |
| Artificial Analysisreported | GPQA Diamond% of questions | 40.9% | 57.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 4.5% | 3.8% |
| Artificial Analysisreported | IFBench% of instructions | 34.4% | 23.5% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 3.0% | 3.8% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 15.2% | 0.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 15.0% | 19.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.38 | 0.34 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 32.8 | 25.7 |
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 3.1 70B Instruct better than Phi 4?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Llama 3.1 70B Instruct ranks higher for writing; Phi 4 ranks higher for reasoning and knowledge.
Which is cheaper, Llama 3.1 70B Instruct or Phi 4?
Phi 4 costs $0.088 per million tokens (three input to one output), against $0.40 for Llama 3.1 70B Instruct.
Run Llama 3.1 70B Instruct and Phi 4 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.