Compare/K2-V2 (medium) vs o3

K2-V2 (medium)vso3

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

MBZUAI Institute of Foundation Models

K2-V2 (medium)

Input
$0/M
Output
$0/M
Speed
TTFT
OpenAI

o3

Input
$2/M
Output
$8/M
Speed
125 tok/s
TTFT
6.65s

Winner by Category

Cheaper
K2-V2 (medium)
Faster (tok/s)
o3
Lower Latency
K2-V2 (medium)
Benchmarks (0-11)
o3

Pricing Comparison

MetricK2-V2 (medium)o3
Input ($/M tokens)$0$2
Output ($/M tokens)$0$8
Cost for 1M input + 100K output tokens:
K2-V2 (medium)$0.00
o3$2.80

Speed Comparison

Output Speed (tokens/s) — higher is better
K2-V2 (medium)
o3
125 tok/s
Time to First Token (seconds) — lower is better
K2-V2 (medium)
o3
6.65s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
12.430.4
Coding Index
Math Index
64.788.3
GPQA Diamond
59.8%82.7%
MMLU-Pro
76.1%85.3%
LiveCodeBench
54.1%80.8%
AIME 2025
64.7%88.3%
MATH-500
99.2%
Humanity's Last Exam
4.4%20.0%
SciCode
25.2%41.0%
IFBench
55.1%71.4%
TerminalBench
8.3%37.1%
K2-V2 (medium)0 wins
11 winso3

Frequently Asked Questions

Which is cheaper, K2-V2 (medium) or o3?

K2-V2 (medium) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.00/M tokens vs $3.50/M for o3.

Which model performs better on benchmarks?

o3 wins 11 out of 12 benchmarks compared to 0 for K2-V2 (medium). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

o3 generates tokens faster at 125 tok/s vs 0 tok/s. K2-V2 (medium) also has lower time-to-first-token (0.00s vs 6.65s).

When should I use K2-V2 (medium) vs o3?

Choose based on your priorities: K2-V2 (medium) for lower cost, o3 for stronger benchmark performance, and o3 for faster generation. For latency-sensitive apps, check the TTFT comparison above.