Run
run-20261102-080410-d2c89897
verdict
not evaluated
score
not measured
time
15 m 2 s
input tokens
989,571
generated tokens
18,787
tokens
1,008,358
tool calls
51
status
timeout
Where this run belongs
| field | value |
|---|---|
| experiment | router-config-comparison@1.0.0 |
| task | slugify-001@1.0.0 |
| parameters | knowledge_enabled=False, model=qwen/Qwen3.8-27B |
| variant | v-5dfe3345dc0f |
| agent | codex@codex-cli 0.166.0 |
| runtime | llm-router |
| mode | router |
| repetition | 2 |
| session | 01a19b49-e012-7623-88bd-7feba182f2a8 |
| manifest digest | sha256:4247f2256b6cb852b122f91ecb10fb39199dee2ac1d84cd6a63772bf6c5418fa |
Evaluation
this run was never evaluated: it has no verdict and no score
What the agent reported
| field | value |
|---|---|
| started | 2026-11-02T08:04:10.026771 |
| finished | 2026-11-02T08:19:12.099648 |
| wall clock | 15 m 2 s |
| input tokens | 989,571 |
| total tokens | 1,008,358 |
| tool calls | 51 |
| exit code | 1 |
| failure class | agent |
| error | I'm looking at the test results more carefully now—most of them are passing, but there are two specific cases where the hyphen handling is producing unexpected triple hyphens instead of the expected output. The issue seems to be that when I'm building the test strings with the hyphen character, something about how the input is being processed is causing the hyphens to multiply in the output. |
Model
| requested | observed | source | verification |
|---|---|---|---|
| qwen/Qwen3.8-27B | qwen/Qwen3.8-27B | agent_reported | unverified |
Plugins observed
| plugin | version | enabled | verification | note |
|---|---|---|---|---|
| agentic_knowledge | 1.0.0 | no | unsupported | LLM_ROUTER_KNOWLEDGE_AGENTIC_ENABLED is not set, so the plugin keeps its built-in default; verification methods effective_router_config, plugin_runtime_events, … |
| agentic_routing_codex | 1.0.0 | no | unsupported | descriptor declares no plugin-level toggle (benchmark.enabled_when), so activation is not observable and enabled is reported as false; verification methods resp … |
Artifacts
| kind | path | checksum |
|---|---|---|
| manifest | /mnt/data2/dev/develop/ak-bench-data/artifacts/runs/codex-20261102-080410-025754-ef4cdd/manifest.json | sha256:48be7dd8a8ad184e90afc72f3a509877f99a8c6ca064de72241832fef6d63411 |
| run_record | /mnt/data2/dev/develop/ak-bench-data/artifacts/runs/codex-20261102-080410-025754-ef4cdd/run_record.json | sha256:ca24836cd54f87dbc7de10b6e2dd72422e1d5c7e5d533a8f78a7e1140db34e7f |
Not verified
Configurations that are not verified. They stay in the report, because a result produced under them is only as trustworthy as they are.
| kind | subject | status | run | note |
|---|---|---|---|---|
| plugin | agentic_knowledge | unsupported | run-20261102-080410-d2c89897 | LLM_ROUTER_KNOWLEDGE_AGENTIC_ENABLED is not set, so the plugin keeps its built-in default; verification methods effective_router_config, plugin_runtime_events, … |
| plugin | agentic_routing_codex | unsupported | run-20261102-080410-d2c89897 | descriptor declares no plugin-level toggle (benchmark.enabled_when), so activation is not observable and enabled is reported as false; verification methods resp … |
| model | qwen/Qwen3.8-27B | unverified | run-20261102-080410-d2c89897 |