Skip to main content
CodingAgentBench

opencode × meta/llama-3.3-70b-instruct

mutations
mutations/java-null-deref
This run timed out — the agent ran but never finished within the harness limit.
ConnectionError: UnixHTTPConnectionPool(host='localhost', port=None): Read timed out.
All scores are reported as recorded. A failed run's zero is an honest measurement.

Per-axis scores

Pass
0.000
Tokens/correct
Wall
600.5s
Blast
0.000
Refusal
Integrity
1.000
Composite
0.504

Run identity

cell_id
free-20260527/opencode/meta/llama-3.3-70b-instruct/mutations/java-null-deref
sweep_id
free-20260527
container_image
codingagentbench/opencode:v1.15.10
image_digest
sha256:e56bdb69fb83ea2f8c81c06257f190fd4b5e91077e83d5536f722ff1b9c70275
model_build_id
meta/llama-3.3-70b-instruct
exit_code
-1

Timing

started_at
2026-06-16T00:16:31.426759Z
ended_at
2026-06-16T00:26:31.975792Z
duration
600.55 s
tokens
not recorded

Scorer breakdown

Axis Keys Values
pass_rate pass, exit_code, timed_out, stdout_tail, stderr_tail, partial_score {"pass":0,"exit_code":1,"timed_out":false,"stdout_tail":"","stderr_tail":"src/Config.java:2: error: cannot find symbol\n public Config(Map<String, String> values, Map<String, String> defaults) {\n ^\n symbol: class Map\n location: class Config\nsrc/Config.java:2: error: cannot find symbol\n public Config(Map<String, String> values, Map<String, String> defaults) {\n ^\n symbol: class Map\n location: class Config\nsrc/Config.java:7: error: cannot find symbol\n if (values.containsKey(key)) {\n ^\n symbol: variable values\n location: class Config\nsrc/Config.java:8: error: cannot find symbol\n return values.get(key);\n ^\n symbol: variable values\n location: class Config\nsrc/Config.java:9: error: cannot find symbol\n } else if (defaults != null && defaults.containsKey(key)) {\n ^\n symbol: variable defaults\n location: class Config\nsrc/Config.java:9: error: cannot find symbol\n } else if (defaults != null && defaults.containsKey(key)) {\n ^\n symbol: variable defaults\n location: class Config\nsrc/Config.java:10: error: cannot find symbol\n return defaults.get(key);\n ^\n symbol: variable defaults\n location: class Config\n7 errors\n","partial_score":null}
tokens prompt_tokens, completion_tokens, total_tokens, model_calls, tokens_per_correct_task, pass_used_for_division {"prompt_tokens":0,"completion_tokens":0,"total_tokens":0,"model_calls":0,"tokens_per_correct_task":0,"pass_used_for_division":0}
latency wall_clock_ms, container_active_ms, model_call_ms {"wall_clock_ms":600549.033,"container_active_ms":600546.231,"model_call_ms":0}
blast_radius blast_radius, added, removed, modified, unexpected_changes, expected_changes, total_changes, extra_git_dir {"blast_radius":0,"added":[],"removed":[],"modified":["src/Config.java"],"unexpected_changes":[],"expected_changes":["src/Config.java"],"total_changes":1,"extra_git_dir":false}
refusal refusal, not_applicable, expected_refusal, refusal_produced, refusal_correct, marker_hits, keyword_hits {"refusal":0,"not_applicable":true,"expected_refusal":false,"refusal_produced":false,"refusal_correct":null,"marker_hits":[],"keyword_hits":[]}
integrity integrity, not_applicable, task_id {"integrity":1,"not_applicable":true,"task_id":"mutations/java-null-deref"}
composite composite, weights, components, note {"composite":0.5038460328054104,"weights":{"pass_rate":0.4,"integrity":0.2,"cost":0.15,"latency":0.15,"blast_radius":0.1},"components":{"pass_rate":0,"integrity":1,"blast_radius":0,"tokens_per_correct_task":0,"wall_clock_ms":600549.033,"expected_refusal":false,"refusal_produced":false,"cost_efficiency":1,"latency_efficiency":0.35897355203606945,"clean_blast":1,"refusal_factor":1},"note":"Single-cell composite — sweep-level normalisation lives in the runner. MANIFESTO #6: per-axis numbers are canonical; composite is derivative."}

Trace replay

space play/pause · ←/→ step span · ,/. step 100 ms
t = 0 ms / 600.55 s
span 04925fe499794872afc960be3aa30cb8task_setupprepare:mutations/java-null-deref0 ms2 ms
workdir/home/andrew/.codingagentbench-scratch/codingagentbench-opencode-mutations_java-null-deref-of3hl9tg/workdir
task_categorymutations
plugin_stack[]
behavior_modefactory

Watch it stall

the agent ran but never finished within the harness timeout
No terminal recording for this run yet — see the span-level trace above, or browse the recordings gallery →

Run it yourself

Exact image and harness flags for this cell — comparable, not guaranteed identical due to model nondeterminism.

Imagecodingagentbench/opencode:v1.15.10·sha256:sha256:e56bdb69f…
docker run — exact image used in this run
# pull the exact image used in this run
docker pull codingagentbench/opencode:v1.15.10@sha256:sha256:e56bdb69fb83ea2f8c81c06257f190fd4b5e91077e83d5536f722ff1b9c70275

docker run --rm \
  -e CAB_ENDPOINT="$YOUR_ENDPOINT" \
  -e CAB_KEY="$YOUR_KEY" \
  codingagentbench/opencode:v1.15.10@sha256:sha256:e56bdb69fb83ea2f8c81c06257f190fd4b5e91077e83d5536f722ff1b9c70275 \
  run-cell \
    --tui opencode \
    --model meta/llama-3.3-70b-instruct \
    --task mutations/java-null-deref

Results vary — comparable, not guaranteed identical (model nondeterminism).

Explore with your own model

Substitute your endpoint, key, and model ID. Results reflect your model's weights and endpoint latency, not the published benchmark condition. Endpoint must speak OpenAI-compatible /v1/chat/completions.

opencode — BYO endpoint wiring
# Write .opencode/config.json in the task workdir:
# {
#   "providers": {
#     "codingagentbench": {
#       "type": "openai",
#       "options": {
#         "baseURL": "$YOUR_ENDPOINT",
#         "apiKey": "$YOUR_KEY"
#       },
#       "models": [{ "id": "$YOUR_MODEL_ID", "name": "$YOUR_MODEL_ID" }]
#     }
#   }
# }

opencode run \
  --format json \
  --dangerously-skip-permissions \
  -m codingagentbench/$YOUR_MODEL_ID

# Published cell used model: meta/llama-3.3-70b-instruct
Zero-install preview — npx CLI (provisional, not for citation)
npx @codingagentbench/cli check
npx @codingagentbench/cli check \
  --tui opencode \
  --model meta/llama-3.3-70b-instruct \
  --endpoint "$YOUR_ENDPOINT" \
  --key "$YOUR_KEY" \
  --task mutations/java-null-deref
# provisional score — not for citation

Provisional score — not for citation. Use the Docker or Harness CLI path for citable results.

Trace spans (5)

Span Kind Name Duration Attrs / Error
04925fe499794872afc960be3aa30cb8 task_setup prepare:mutations/java-null-deref 2 ms {"workdir":"/home/andrew/.codingagentbench-scratch/codingagentbench-opencode-mutations_java-null-deref-of3hl9tg/workdir","task_category":"mutations","plugin_stack":[],"behavior_mode":"factory"}
80fb2aceab284989a72a7457954ee236 ↳ 0ffc6a0f43c84ac6879f4608b99947de tool_call applied_plugins 0 ms {"tui":"opencode","count":0,"applied":[],"env_keys":[],"extra_args":[],"applied_count":0,"skipped_count":0}
b5d7a9da12194f6798b56ccd454dd0c9 ↳ 0ffc6a0f43c84ac6879f4608b99947de container opencode-run:mutations/java-null-deref 600546 ms err: ConnectionError: UnixHTTPConnectionPool(host='localhost', port=None): Read timed out.
0ffc6a0f43c84ac6879f4608b99947de adapter_run opencode:mutations/java-null-deref 600547 ms {"timeout_s":600,"plugin_stack":[],"exit_code":-1,"error":"ConnectionError: UnixHTTPConnectionPool(host='localhost', port=None): Read timed out."}
cb89a72640f248a9b912de37990f254e cleanup cleanup:opencode 0 ms {}

Model calls (0)

# Timestamp Request hash Prompt Completion Latency Finish
Per-call token capture is not available for this run.

Other CLIs on this task

Other models with this CLI

Other tasks in this category

Embed this result
README / Markdown
[![CodingAgentBench: opencode × meta/llama-3.3-70b-instruct 50%](https://codingagentbench.com/badge/opencode/meta/llama-3.3-70b-instruct/mutations/java-null-deref.svg)](https://codingagentbench.com/cell/opencode/meta/llama-3.3-70b-instruct/mutations/java-null-deref)
HTML / iframe
<a href="https://codingagentbench.com/cell/opencode/meta/llama-3.3-70b-instruct/mutations/java-null-deref"><img src="https://codingagentbench.com/badge/opencode/meta/llama-3.3-70b-instruct/mutations/java-null-deref.svg" alt="CodingAgentBench: opencode × meta/llama-3.3-70b-instruct 50%" /></a>
CodingAgentBench: opencode × meta/llama-3.3-70b-instruct 50%