Compare/Kimi K2 0905 vs GLM-4.6 (Reasoning)

Kimi K2 0905vsGLM-4.6 (Reasoning)

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

Kimi

Kimi K2 0905

Input
$0.8/M
Output
$2.25/M
Speed
63 tok/s
TTFT
0.78s
Z AI

GLM-4.6 (Reasoning)

Input
$0.575/M
Output
$2.2/M
Speed
84 tok/s
TTFT
0.67s

Winner by Category

Cheaper
GLM-4.6 (Reasoning)
Faster (tok/s)
GLM-4.6 (Reasoning)
Lower Latency
GLM-4.6 (Reasoning)
Benchmarks (0-11)
GLM-4.6 (Reasoning)

Pricing Comparison

MetricKimi K2 0905GLM-4.6 (Reasoning)
Input ($/M tokens)$0.8$0.575
Output ($/M tokens)$2.25$2.2
Cost for 1M input + 100K output tokens:
Kimi K2 0905$1.03
GLM-4.6 (Reasoning)$0.79

Speed Comparison

Output Speed (tokens/s) — higher is better
Kimi K2 0905
63 tok/s
GLM-4.6 (Reasoning)
84 tok/s
Time to First Token (seconds) — lower is better
Kimi K2 0905
0.78s
GLM-4.6 (Reasoning)
0.67s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
30.932.5
Coding Index
25.929.5
Math Index
57.386.0
GPQA Diamond
76.7%78.0%
MMLU-Pro
81.9%82.9%
LiveCodeBench
61.0%69.5%
AIME 2025
57.3%86.0%
MATH-500
Humanity's Last Exam
6.3%13.3%
SciCode
30.7%38.4%
IFBench
41.7%43.4%
TerminalBench
23.5%25.0%
Kimi K2 09050 wins
11 winsGLM-4.6 (Reasoning)

Frequently Asked Questions

Which is cheaper, Kimi K2 0905 or GLM-4.6 (Reasoning)?

GLM-4.6 (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.98/M tokens vs $1.14/M for Kimi K2 0905.

Which model performs better on benchmarks?

GLM-4.6 (Reasoning) wins 11 out of 12 benchmarks compared to 0 for Kimi K2 0905. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GLM-4.6 (Reasoning) generates tokens faster at 84 tok/s vs 63 tok/s. However, GLM-4.6 (Reasoning) has lower time-to-first-token (0.67s vs 0.78s).

When should I use Kimi K2 0905 vs GLM-4.6 (Reasoning)?

Choose based on your priorities: GLM-4.6 (Reasoning) for lower cost, GLM-4.6 (Reasoning) for stronger benchmark performance, and GLM-4.6 (Reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.