mlsmaycon c8eb604f35 Run a single live Kimi scenario: kimi-k3 over the Anthropic shape ...

The Moonshot platform serves this account exactly one model — kimi-k3.
kimi-k2-thinking and the kimi-latest alias both return
resource_not_found_error on either API surface, so two concurrent Kimi
provider records would claim the same model and route ambiguously
(the router's path-prefix tiebreak needs the client to author the
/anthropic prefix, which the standard wire shapes don't). Keep the
flagship Claude Code path — kimi-k3 via /anthropic, green in run #74 —
as the live scenario; the OpenAI wire shape stays covered by the other
chat-shaped matrix providers and passed with kimi-k3 in run #73.

2026-07-21 13:02:03 +00:00

..

bootstrap_test.go

[management,proxy,client] 0.74.0 version (#6563)

2026-07-01 12:45:14 +02:00

chat_test.go

Run a single live Kimi scenario: kimi-k3 over the Anthropic shape

2026-07-21 13:02:03 +00:00

guardrail_test.go

Use the full Bedrock inference-profile id in the e2e matrix

2026-07-21 11:49:42 +00:00

main_test.go

[management,proxy,client] 0.74.0 version (#6563)

2026-07-01 12:45:14 +02:00

management_test.go

[management,proxy,client] 0.74.0 version (#6563)

2026-07-01 12:45:14 +02:00

skiptls_test.go

[management,proxy] Add per-provider skip_tls_verification for agent-network (#6630)

2026-07-01 20:43:15 +02:00

vllm_test.go

[management] Add vLLM e2e test (#6649)

2026-07-02 15:36:51 +02:00