provider-scoped/openrouter/moonshotai/kimi-k2.7-code:batch
Source-backed model information from the Foundry registry.
Agent skill
The active tab is a live, model-specific SKILL.md. Copy it into the harness or fetch the raw file at the canonical URL.
Overview
No source-backed public description is present in this registry snapshot.
Pricing
| Token type | Rate |
|---|---|
| Input tokens | — |
| Output tokens |
| — |
| Cache read | — |
|---|
| Cache write | — |
|---|
Rates are rendered only when published by the current live Registry Intelligence snapshot.
Model details
API
Choose this model in the API tab to generate current, copyable directions from the live model and provider registry.
Authenticated preview
Playground requests use Foundry's authenticated completions path. Connect an API key to run a real request.
No request is sent and no assistant response is simulated here.
Connect an API keyServing paths
Routes for this model, with the originating lab kept first.
| Provider | Context | Max output | Latency | Throughput | Input | Output | Cache read | State |
|---|---|---|---|---|---|---|---|---|
| OpenRoutermoonshotai/kimi-k2.7-code:batch | 262,144 | 235,929 | — | — | $0.95 / 1M | $4 / 1M | $0.19 / 1M | Unavailable |
| OpenRoutermoonshotai/kimi-k2.7-code:batch | 262,144 | 235,929 | — | — | $0.95 / 1M | $4 / 1M | $0.19 / 1M | Unavailable |
Direct request success rate on Foundry and per-provider. Visit the docs ↗
Foundry only renders measurements published by the external telemetry machine. No live values are being invented in this page.
FOUNDRY_TELEMETRY_URLReliability of providers. Visit the docs ↗
Foundry only renders measurements published by the external telemetry machine. No live values are being invented in this page.
FOUNDRY_TELEMETRY_URLP50 throughput on live Foundry traffic, in tokens per second (TPS). Visit the docs ↗
Foundry only renders measurements published by the external telemetry machine. No live values are being invented in this page.
FOUNDRY_TELEMETRY_URLP50 time to first token (TTFT) on live Foundry traffic, in milliseconds. Visit the docs ↗
Foundry only renders measurements published by the external telemetry machine. No live values are being invented in this page.
FOUNDRY_TELEMETRY_URLGetting started
Use the agent skill at the top of this page to configure the selected harness, refresh the live registry, and call OpenAI-compatible chat completions. Foundry authenticates the request and routes it to an available provider.
Live integration
Select a model from the current live registry. Foundry emits the harness binding, API key setup, live provider context, chat completions request, and fail-closed error handling as one copyable recipe.
Related models
Rows are selected from the current live catalog.
| Model | Context | Latency | Throughput | Input | Output | Cache | Capabilities |
|---|---|---|---|---|---|---|---|
| moonshotai/kimi-k2.7-code | — | — | — | — | — | — | chatchat_completionimage_understanding+2 |
| moonshotai/kimi-k2.7-code:batch | — | — | — | — | — | — | chatchat_completionimage_understanding+2 |
| moonshotai/kimi-k2.7-code-highspeed | — | — | — | — | — | — | chatchat_completiontext_generation |