Analytics

Which model fits your task

Price, speed and quality on real traffic over 14 days. Updated hourly.

We recommend

Claude Sonnet 4.5

Handles long IDE tasks better than anyone and is careful with tools.

Per answer
$0.011
Speed
71 tok/s
Quality
91
Cheapest

$0.00013 / answer

gpt-oss-120b

Fastest

190 tok/s

Gemini 2.5 Flash

Smartest

92 / 100

GPT-5

First token

0.4 s

Gemini 2.5 Flash

Answer price

Lower is better

  • gpt-oss-120b
    $0.00013
  • Llama 3.3 70B
    $0.00028
  • DeepSeek V3.1
    $0.00077
  • Qwen3 Coder
    $0.00085
  • GPT-5 mini
    $0.0013
  • Gemini 2.5 Flash
    $0.0016
  • Kimi K2.6
    $0.0026
  • Claude Haiku 4.5
    $0.0035
  • GPT-5
    $0.0063
  • Gemini 2.5 Pro
    $0.0063
  • Claude Sonnet 4.5
    $0.011

Price ↔ quality

Top-left corner is the best deal

808590950$0.0036$0.0072$0.011GPT-5Claude Sonnet 4.5Gemini 2.5 ProDeepSeek V3.1Qwen3 CoderGPT-5 miniClaude Haiku 4.5Gemini 2.5 Flashgpt-oss-120bKimi K2.6Llama 3.3 70B

Time to first token, 14 days

Median across all requests, seconds

00.81.62.4Sep 11Sep 13Sep 15Sep 17Sep 19Sep 21Sep 23

Head to head

Pick two models — the winner on each metric is highlighted

Price per answer$0.011$0.00077 ✓
Input / 1M$3.00$0.27 ✓
Output / 1M$15.00$1.00 ✓
Speed71 tok/s ✓58 tok/s
First token1.1 s ✓1.6 s
Quality91 ✓84
Context1M160K