Model comparison
Gemini 3.1 Pro vs Kimi K3
Gemini
KimiCompare published specifications, vendor API list prices and measured Borg performance. Both models are included in Borg.
Gemini 3.1 Pro
Kimi K3
Included in Borg
At a glance
| Specification | Gemini 3.1 Pro | Kimi K3 |
|---|---|---|
| Provider | Moonshot AI | |
| Release | 2026-02-19 Vendor source | 2026-07-16 Vendor source |
| Context | 1,048,576 tokens Vendor source | 1,048,576 tokens Vendor source |
| Modalities | Input: text, image, video, audio, pdf; output: text | Input: text, image, video; output: Not published |
| Reasoning / thinking modes | Yes / Not published | Yes / Not published |
| Open weights | No | Yes |
| Vendor API list price / 1M tokens | $2 input / $12 output Rates for prompts of at most 200,000 tokens; longer prompts cost more. | $3 input / $15 output |
| Borg | Included in Borg subscription | Included in Borg subscription |
Google
Moonshot AIMeasured 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)
No published measurement for this model.
Output tokens per second (whole turn)
No published measurement for this model.
| Metric | Gemini 3.1 Pro | Kimi K3 |
|---|---|---|
| Time to first token (gateway proxy) p50 | Not published | 8627 ms (7633 samples) |
| Tokens per second (end to end) p50 | Not published | 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.
Gemini 3.1 Pro: Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 278 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
Gemini 3.1 Pro is listed for complex software engineering, whereas Kimi K3 is listed for long-horizon coding across large repositories. Choosing between Gemini 3.1 Pro and Kimi K3 requires checking published specifications against observed Borg traffic for your workflow.
Key differences
| Spec | Gemini 3.1 Pro | Kimi K3 |
|---|---|---|
| Context window | 1,048,576 tokens Vendor source | 1,048,576 tokens Vendor source |
| Reasoning modes | Yes / Not published | Yes / Not published |
| Release date | 2026-02-19 Vendor source | 2026-07-16 Vendor source |
| API price input / 1M | 2 Rates for prompts of at most 200,000 tokens; longer prompts cost more. Lower listed rate | 3 |
| API price output / 1M | 12 Rates for prompts of at most 200,000 tokens; longer prompts cost more. Lower listed rate | 15 |
| Time to first token (gateway proxy) | Not published | 8627 ms (7633 samples) |
| Output tokens per second (whole turn) | Not published | 18.1 output tokens/s (whole turn) (152924 samples) |
| Request success | Not published | Not published |
| Coding bench | Not published | Not published |
FAQ
How do Gemini 3.1 Pro and Kimi K3 compare for coding?
Gemini 3.1 Pro has a Borg coding benchmark of not published; Kimi K3 has not published. The Gemini 3.1 Pro–Kimi K3 coding comparison cannot establish an overall winner from this small test alone.
How quickly do Gemini 3.1 Pro and Kimi K3 start responding on Borg?
Borg median first response for Gemini 3.1 Pro is not published; for Kimi K3 it is 8,627 ms. The Gemini 3.1 Pro–Kimi K3 gateway medians are observations, not guaranteed latency for individual requests.
What are the API prices for Gemini 3.1 Pro and Kimi K3?
The listed API rate for Gemini 3.1 Pro is $2 input / $12 output per 1M tokens; for Kimi K3 it is $3 input / $15 output per 1M tokens. Neither Gemini 3.1 Pro nor Kimi K3 has its Borg subscription charge represented by those provider API rates. Gemini 3.1 Pro vendor source · Kimi K3 vendor source
Claude