R
Relayby Ai5labs
← Back to all models

Compare models

Comparing 2models. Drop the URL into a doc — it's permalinked.

Fieldgoogle/gemini-2.5-flashdeepseek/deepseek-chat
Providergoogledeepseek
Model IDgemini-2.5-flashdeepseek-chat
Context1.0M164K
Max output8K8K
Input / 1M$0.30$0.32
Output / 1M$2.50$0.89
Cached input / 1M$0.07$0.07
Avg cost / 1M$1.40$0.60
Speed250 t/s50 t/s
Quality index70.072.0
MMLU85.387.1
GPQA65.259.1
HumanEval88.189.0
MATH88.090.2
SWE-bench
Arena Elo
Tools
Vision
Thinking
Streaming
JSON mode
Structured output
Prompt cache

Same data, in your terminal: relay models compare gemini-flash deepseek-v3

Suggested comparisons