kimi-k3 vs glm-5.2
Side-by-side comparison of kimi-k3 and glm-5.2: benchmarks, pricing, context window and capabilities. Both are accessible through Requesty's unified API. kimi-k3 outperforms glm-5.2 on 5 of 5 shared benchmarks.
kimi-k3
Input / 1M
$3.00
Output / 1M
$15.00
Context
1.0M
Model ID
moonshot/kimi-k3
glm-5.2
Input / 1M
$1.40
Output / 1M
$4.40
Context
1M
Model ID
zai/glm-5.2
Benchmark comparison
Intelligence Indexreasoning
kimi-k350.2%
glm-5.242.1%
Coding Indexcoding
kimi-k376.2%
glm-5.268.8%
GPQA Diamondreasoning
kimi-k393.5%
glm-5.289.5%
Terminal-Bench Hardagentic
kimi-k3N/A
glm-5.250.8%
τ²-Benchagentic
kimi-k3N/A
glm-5.299.1%
SciCodecoding
kimi-k359.5%
glm-5.251.2%
Humanity's Last Examreasoning
kimi-k346.9%
glm-5.241.1%
Scores sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and don't capture every aspect of model quality.
Pricing & specifications
| kimi-k3 | glm-5.2 | |
|---|---|---|
| Input price / 1M | $3.00 | $1.40 |
| Output price / 1M | $15.00 | $4.40 |
| Context window | 1.0M tokens | 1M tokens |
| Max output | 1.0M tokens | 128K tokens |
| Vision input | Yes | N/A |
| Tool calling | Yes | Yes |
| Reasoning | Yes | Yes |
| Prompt caching | Yes | Yes |
| Computer use | N/A | N/A |
| Provider | Moonshot AI | Z.ai |
Questions people ask
Is kimi-k3 better than glm-5.2?
kimi-k3 outperforms glm-5.2 on 5 of 5 shared benchmarks. See the benchmark comparison above for specifics: kimi-k3 and glm-5.2 have different strengths across reasoning, coding, math and multimodal tasks.
Which is cheaper, kimi-k3 or glm-5.2?
glm-5.2 is cheaper. kimi-k3 costs $3.00/$15.00 per 1M input/output tokens, while glm-5.2 costs $1.40/$4.40.
Can I use kimi-k3 and glm-5.2 through the same API?
Yes. Requesty provides a single OpenAI-compatible API that routes to both. Change just the "model" parameter to switch between "moonshot/kimi-k3" and "zai/glm-5.2", no other code changes needed.
What are the context windows?
kimi-k3 supports up to 1.0M tokens of context. glm-5.2 supports up to 1M tokens. Longer context means you can feed larger documents or codebases in a single prompt, though quality often degrades past 128K for most models.
Switch between kimi-k3 and glm-5.2 with one line of code
Requesty provides a single OpenAI-compatible API for 600+ models. Change the model parameter, not your code.