Phi-4: Intelligence, Performance & Price Analysis
Analysis of Microsoft Azure's Phi-4 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-4 is of higher quality compared to average, with a MMLU score of 0.714 and a Intelligence Index across evaluations of 40.
Price:Phi-4 is cheaper compared to average with a price of $0.22 per 1M Tokens (blended 3:1).
Phi-4 Input token price: $0.13, Output token price: $0.50 per 1M Tokens.
Speed:Phi-4 Input token price: $0.13, Output token price: $0.50 per 1M Tokens.
Phi-4 is slower compared to average, with a output speed of 23.8 tokens per second.
Latency:Phi-4 has a lower latency compared to average, taking 0.46s to receive the first token (TTFT).
Context Window:Phi-4 has a smaller context windows than average, with a context window of 16k tokens.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Phi-4 Model Details
Comparisons to Phi-4
- Phi-4
o3- o4-mini (high)
- GPT-4.1
- o3-pro
Llama 4 Maverick- Llama 4 Scout
Gemini 2.5 Flash (Reasoning)- Gemini 2.5 Pro
Claude 4 Opus Thinking- Claude 4 Sonnet Thinking
Magistral Small
DeepSeek V3 0324 (Mar '25)- DeepSeek R1 0528 (May '25)
Grok 3 mini Reasoning (high)- Grok 4
Nova Premier
MiniMax M1 80k
Llama Nemotron Ultra Reasoning
Qwen3 235B (Reasoning)- GPT-4o (Nov '24)
Further details
Sonar
LFM 40B
Solar Pro 2 (Preview) (Reasoning)
DBRX Instruct
Llama 3.1 Tulu3 405B
Reka Flash (Sep '24)
Hermes 3 - Llama-3.1 70B
Aya Expanse 32B
Jamba 1.6 Large
Arctic Instruct
Yi-Large
OpenChat 3.5 (1210)