Model comparison
Kimi K2.7 Code vs Kimi K3
KimiCompare published specifications, vendor API list prices and measured Borg performance. Both models are included in Borg.
Kimi K2.7 Code
Kimi K3
Included in Borg
At a glance
| Specification | Kimi K2.7 Code | Kimi K3 |
|---|---|---|
| Provider | Moonshot AI | Moonshot AI |
| Release | — | 2026-07-16 Vendor source |
| Context | 262,144 tokens Vendor source | 1,048,576 tokens Vendor source |
| Modalities | Input: text, image, video; output: Not published | Input: text, image, video; output: Not published |
| Reasoning / thinking modes | Yes / Not published | Yes / Not published |
| Open weights | Yes | Yes |
| Vendor API list price / 1M tokens | $0.95 input / $4 output | $3 input / $15 output |
| Borg | Included in Borg subscription | Included in Borg subscription |
Moonshot AINot published by the vendor: release.
Measured on Borg
Metrics are not directly comparable to vendor benchmark claims. Models served different requests; this is observed gateway traffic, not a controlled head-to-head test.
Time to first token (gateway proxy)
Output tokens per second (whole turn)
| Metric | Kimi K2.7 Code | Kimi K3 |
|---|---|---|
| Time to first token (gateway proxy) p50 | 4780 ms (119 samples) | 8627 ms (7633 samples) |
| Tokens per second (end to end) p50 | 24.2 output tokens per second (1894 samples) | 18.1 output tokens per second (152924 samples) |
| Request success rate | Not published | Not published |
| Coding bench pass rate | Not published | Not published |
Measurement window, samples and methodology
The fleet reference is the median of published model medians, not a request-weighted result. Prompt lengths, reasoning settings and traffic differ between models.
Kimi K2.7 Code: Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 2583 requests. Minimum 100 samples per published request metric.
Kimi K3: Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 183015 requests. Minimum 100 samples per published request metric.
Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.
- Time to first token (gateway proxy): time until the first streamed response event reaches the gateway; our closest available measure of time to first token
- Tokens per second (end to end): tokens per second, end to end: provider-reported output tokens divided by the whole turn's elapsed time, including queueing and reasoning; not a streaming per-token rate
How to choose
Kimi K2.7 Code is listed for multi-file coding with iterative test runs, whereas Kimi K3 is listed for long-horizon coding across large repositories. Choosing between Kimi K2.7 Code and Kimi K3 requires checking published specifications against observed Borg traffic for your workflow.
Key differences
| Spec | Kimi K2.7 Code | Kimi K3 |
|---|---|---|
| Context window | 262,144 tokens Vendor source | 1,048,576 tokens Vendor source Larger context |
| Reasoning modes | Yes / Not published | Yes / Not published |
| Release date | Not published | 2026-07-16 Vendor source |
| API price input / 1M | 0.95 Lower listed rate | 3 |
| API price output / 1M | 4 Lower listed rate | 15 |
| Time to first token (gateway proxy) | 4780 ms (119 samples) Lower median | 8627 ms (7633 samples) |
| Output tokens per second (whole turn) | 24.2 output tokens/s (whole turn) (1894 samples) Higher measured rate | 18.1 output tokens/s (whole turn) (152924 samples) |
| Request success | Not published | Not published |
| Coding bench | Not published | Not published |
FAQ
How do Kimi K2.7 Code and Kimi K3 compare for coding?
Kimi K2.7 Code has a Borg coding benchmark of not published; Kimi K3 has not published. The Kimi K2.7 Code–Kimi K3 coding comparison cannot establish an overall winner from this small test alone.
How quickly do Kimi K2.7 Code and Kimi K3 start responding on Borg?
Borg median first response for Kimi K2.7 Code is 4,780 ms; for Kimi K3 it is 8,627 ms. The Kimi K2.7 Code–Kimi K3 gateway medians are observations, not guaranteed latency for individual requests.
What are the API prices for Kimi K2.7 Code and Kimi K3?
The listed API rate for Kimi K2.7 Code is $0.95 input / $4 output per 1M tokens; for Kimi K3 it is $3 input / $15 output per 1M tokens. Neither Kimi K2.7 Code nor Kimi K3 has its Borg subscription charge represented by those provider API rates. Kimi K2.7 Code vendor source · Kimi K3 vendor source
Claude