Compare · ModelsLive · 2 picked · head to head
Llama 3-70B vs phi-3-small 7.4B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
phi-3-small 7.4B wins on 3/6 benchmarks
phi-3-small 7.4B wins 3 of 6 shared benchmarks. Leads in knowledge.
Category leads
knowledge·phi-3-small 7.4B
Hype vs Reality
Attention vs performance
Llama 3-70B
#257 by perf·#19 by attention
phi-3-small 7.4B
#32 by perf·#20 by attention
Vendor risk
Who is behind the model
Meta AI
$1.65T·Tier 1
Microsoft
$3.96T·Big Tech
Head to head
6 benchmarks · 2 models
Llama 3-70Bphi-3-small 7.4B
ANLI
phi-3-small 7.4B leads by +1.2
ANLI (Adversarial NLI) · adversarially constructed natural language inference dataset where each round targets weaknesses found in previous model generations.
Llama 3-70B
36.0
phi-3-small 7.4B
37.1
ARC AI2
phi-3-small 7.4B leads by +10.5
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
Llama 3-70B
77.1
phi-3-small 7.4B
87.6
MMLU
Llama 3-70B leads by +4.8
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Llama 3-70B
72.4
phi-3-small 7.4B
67.6
OpenBookQA
phi-3-small 7.4B leads by +7.2
OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting.
Llama 3-70B
76.8
phi-3-small 7.4B
84.0
TriviaQA
Llama 3-70B leads by +9.6
TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents.
Llama 3-70B
67.7
phi-3-small 7.4B
58.1
Winogrande
Llama 3-70B leads by +4.0
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Llama 3-70B
67.0
phi-3-small 7.4B
63.0
Full benchmark table
| Benchmark | Llama 3-70B | phi-3-small 7.4B |
|---|---|---|
ANLI ANLI (Adversarial NLI) · adversarially constructed natural language inference dataset where each round targets weaknesses found in previous model generations. | 36.0 | 37.1 |
ARC AI2 AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval. | 77.1 | 87.6 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 72.4 | 67.6 |
OpenBookQA OpenBookQA · science questions that require combining a given core fact with broad common knowledge, mimicking an open-book exam setting. | 76.8 | 84.0 |
TriviaQA TriviaQA · reading comprehension benchmark with trivia questions, requiring models to find and reason over evidence from provided documents. | 67.7 | 58.1 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 67.0 | 63.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| n/a | n/a | n/a | n/a | |
| n/a | n/a | n/a | n/a |
People also compared