Models / Compare /Kimi K2.7 Code vs GLM 5.1
Head to head
Kimi K2.7 Code vs GLM 5.1
Compared on live catalog data — pricing synced daily, benchmark scores with provenance.
The short version
- GLM 5.1 is cheaper to run hosted: $3.04/M output vs $3.50/M for Kimi K2.7 Code (1.2× difference at the cheapest tracked provider).
- Kimi K2.7 Code takes 262K tokens of context vs 205K for GLM 5.1.
- Kimi K2.7 Code has downloadable weights (OPEN*) so you can self-host it; we hold no downloadable copy of GLM 5.1, so it runs through a provider's API.
- GLM 5.1 scores 0.0044 on Arena Agent (IPS) vs 0.0044 for Kimi K2.7 Code.
- GLM 5.1 scores 1516.1626 on Arena Code (WebDev) vs 1472.5712 for Kimi K2.7 Code.
| Spec | ||
|---|---|---|
| Vendor | Moonshot AI | Z.AI |
| Parameters | 1.1T / 32B | 754B |
| 262K | 205K | |
| Weights | OPEN* | OPEN |
| License | Custom licence | MIT License |
| Modality | text+image->text | text->text |
| Released | Jun 11, 2026 | Apr 3, 2026 |
| 667.5 GBest | 475.3 GBest | |
| Cheapest hosted | $0.73 / $3.50 per 1M (OpenRouter) | $0.97 / $3.04 per 1M (OpenRouter) |
| Tracked providers | 20 | 23 |
Kimi K2.7 Code — our take
Moonshot's very large coding model can be downloaded. At over a trillion parameters it is an unusually large model to be downloadable at all, aimed at agentic coding workloads where scale matters.
full breakdown →GLM 5.1 — our take
GLM 5.1 is a large downloadable text model from Z.AI with a permissive MIT licence and a 204,800-token request limit. Its strongest measured skill is coding, though creative writing lags well behind, and no active-parameter count is disclosed.
full breakdown →Need more columns? Use theinteractive comparison to add up to four models.