Phi-3 Mini Instruct 3.8B: Intelligence, Performance & Price Analysis
Analysis of Microsoft Azure's Phi-3 Mini Instruct 3.8B and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more. Click on any model to compare API providers for that model. For more details including relating to our methodology, see our FAQs.
Comparison Summary
Intelligence:
Phi-3 Mini is of lower intelligence compared to average, with Artificial Analysis Intelligence Index of 13.
Price:Phi-3 Mini is cheaper compared to average with a price of $0.23 per 1M Tokens (blended 3:1).
Phi-3 Mini Input token price: $0.13, Output token price: $0.52 per 1M Tokens.
Speed:Phi-3 Mini Input token price: $0.13, Output token price: $0.52 per 1M Tokens.
Phi-3 Mini is slower compared to average, with a output speed of 68.1 tokens per second.
Latency:Phi-3 Mini has a lower latency compared to average, taking 0.35s to receive the first token (TTFT).
Context Window:Phi-3 Mini has a smaller context windows than average, with a context window of 4.1k tokens.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Estimate (independent evaluation forthcoming)
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Note: Long prompts not supported as a context window of at least 10k tokens is required
Phi-3 Mini Instruct 3.8B Model Details
Comparisons to Phi-3 Mini
Phi-3 Mini
gpt-oss-20B (high)
gpt-oss-120B (high)
GPT-5 (minimal)
GPT-4.1
GPT-5 (high)
o3
Llama 4 Maverick
Gemini 2.5 Pro
Gemini 2.5 Flash
Claude 4.1 Opus
Claude 4 Sonnet
Mistral Medium 3.1
DeepSeek R1 0528
DeepSeek V3.1
DeepSeek V3.1
Grok 4
Grok Code Fast 1
Solar Pro 2
Llama Nemotron Super 49B v1.5
Kimi K2 0905
EXAONE 4.0 32B
GLM-4.5
Qwen3 235B 2507
Further details