arm
cnn_deepseek_m_n200_v1 · judged/judge-summ-pick-v1
One config on cnn_dailymail_200, scored one way; its repeats averaged per item.
- dataset
- cnn_dailymail_200
- scored against
- judge
judge-summ-pick-v1 - goal
- Short news summaries (CNN)
- runs
- 1
- headline
- judge_won 0.083
- last run
- 2026-11-02 16:17
01metrics
| metric | mean | items | better |
|---|---|---|---|
| judge_won | 0.083 | 198 | higher |
| cost_usd | $0.000189 | 198 | lower |
| judge_gemma_l | 0.091 | 198 | higher |
| judge_qwen_s | 0.076 | 198 | higher |
| latency_ms | 1592.0 | 198 | lower |
| reasoning_tokens | 0.000 | 198 | neutral |
| tokens_in | 731.3 | 198 | lower |
| tokens_out | 75.6 | 198 | lower |
| truncated | 0.000 | 198 | neutral |
02cost and speed
- cost per item
- $0.000189 · 807 tok
- speed per item
- p50 1,497 · p95 2,927 ms
Per item, over its items: cost as dollars and tokens (a local model is free but still uses tokens), speed as the median and the slow tail (p95). A goal that names this arm weighs them in its decision.
03runs
| run | made | items | cost | how |
|---|---|---|---|---|
cnn_deepseek_m_n200_v1_20260925T233650Z | 2026-11-02 16:17 | 198 | $0.037378 | judge |
04config
config cnn_deepseek_m_n200_v1 · adapter examples/summarization-cnn-dailymail/adapter.py · credentials are never shown
- family
deepseek- max_tokens
1200- model
eval-deepseek-41flash- price_tier
mid- prompt_file
prompt.txt- provider
litellm- reasoning
{"enabled": false}- temperature
0.0- usd_per_mtok_in
0.15- usd_per_mtok_out
0.5
05conditions
- data.items_sha256
257feda902391b934b8cde940c15e91201096dbe96ec5a81337e9da007ce3ec7- deployment.model
eval-deepseek-41flash- deployment.source
adapter- deployment.upstream
openrouter/deepseek/deepseek-v4.1-flash- fingerprint.version
5- library.numpy
2.4.6- library.openai
3.19.2- library.rouge-score
0.1.2- machine.cpu_count
12- machine.platform
Darwin/x86_64- machine.processor
i386- machine.python
3.11.8- machine.release
24.6.0- reference.sha256
5d0b02c3798017535572d24dd3ce0592332938a0c91e93684d163963a60a7cd9- scorer.id
judge-score:judge-summ-pick-v1- scorer.sha256
7046021641ded3f501d1f940ca5cbf5cd06fa66131cbca684f2566f1a031844c
06the same config, scored other ways
scored against silverscored by a judgescored against a label set, not a reference setunmarked rows are scored against gold
| arm | scored against | headline | mean |
|---|---|---|---|
| cnn_deepseek_m_n200_v1 | gold/cnn_dailymail_200 gold | coverage | 0.346 |
| cnn_deepseek_m_n200_v1 · silver | silver/cnn_dailymail_200 silver | coverage | 0.471 |
| cnn_deepseek_m_n200_v1 · judged/judge-summ-pointwise-v1 | judge judge-summ-pointwise-v1 | judge_score_unvalidated | 4.940 |
| cnn_deepseek_m_n200_v1 · judged/judge-summ-v1 | judge judge-summ-v1 | judge_score | 4.923 |
| cnn_deepseek_m_n200_v1 · judged/judge-summ-v2 | judge judge-summ-v2 | judge_score | 4.854 |
07rounds
- Round r1 of Short news summaries (CNN), closed: one of its arms.