Add leaderboard entry: Claude Code / DeepSeek V4 Pro - #11
Open
ajayr4j wants to merge 1 commit into
Open
Conversation
Bare Claude Code harness (no skills, no system-prompt injection) run against deepseek-v4-pro on Enterprise-Bench/l1-l2-bench, 14 tasks x 10 attempts = 140 trials, 1 transient upstream error. Cost/token telemetry: this run routed through litellm to a non-Anthropic model, and Claude Code's stream-json output did not populate real usage stats (fields are 0 or near-0 throughout, even for successful trials) -- the same instrumentation gap the GLM-5.2 entry (devrev#9) hit. Cost uses that entry's "-" display convention (total_cost_usd: 0.0). Token fields (uncached_input_tokens: 41288, output_tokens: 390, total_tokens: 41678) are Harbor's own reported aggregate for this job; they are known to undercount the real 140-trial total (they line up with only the single errored trial, not the full run) and are included as-is because the schema requires numeric values and no complete, independently-verifiable token count was obtainable for this run. Flagging this explicitly for reviewers so the numbers aren't read as a precise measurement. Source job: https://hub.harborframework.com/jobs/2dbebc25-52e9-4ab3-9e6f-c8201c78975a
This was referenced Aug 1, 2026
Author
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add leaderboard entry: Claude Code / DeepSeek V4 Pro
Bare Claude Code harness (no skills, no system-prompt injection) run
against deepseek-v4-pro on Enterprise-Bench/l1-l2-bench, 14 tasks x 10
attempts = 140 trials, 1 transient upstream error.
Cost/token telemetry: this run routed through litellm to a
non-Anthropic model, and Claude Code's stream-json output did not
populate real usage stats (fields are 0 or near-0 throughout, even for
successful trials) -- the same instrumentation gap the GLM-5.2 entry
(#9) hit. Cost uses that entry's "-" display convention
(total_cost_usd: 0.0). Token fields (uncached_input_tokens: 41288,
output_tokens: 390, total_tokens: 41678) are Harbor's own reported
aggregate for this job; they are known to undercount the real 140-trial
total (they line up with only the single errored trial, not the full
run) and are included as-is because the schema requires numeric
values and no complete, independently-verifiable token count was
obtainable for this run. Flagging this explicitly for reviewers so the
numbers aren't read as a precise measurement.
Source job: https://hub.harborframework.com/jobs/2dbebc25-52e9-4ab3-9e6f-c8201c78975a