Add benchmark results for anthropic/claude-opus-4.1 #334
Conversation
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
- Ran 50 benchmark cases with concurrency 15 - Success rate: 58% (29/50 successful) - Total cost: $58.96 - All failures were output mismatches, no API errors Mentat precommit script passed. Log: https://mentat.ai/gh/AbanteAI/LoCoDiff-bench/log/38762b95-d970-4ed0-9002-1799026a60a0 Co-authored-by: biobootloader <128252497+biobootloader@users.noreply.github.com>
- Ran additional 50 benchmark cases (total now 100/200) - Second batch success rate: 60% (30/50 successful) - Combined success rate: 59% (59/100 successful) - Second batch cost: $53.51 - Total cost: $112.47 - All failures were output mismatches, no API errors Mentat precommit script passed. Log: https://mentat.ai/gh/AbanteAI/LoCoDiff-bench/log/ffb63411-4236-49bc-b205-8de2ff3108c6 Co-authored-by: biobootloader <128252497+biobootloader@users.noreply.github.com>
- Completed full benchmark suite: 200/200 cases - Third batch success rate: 62% (62/100 successful) - Final overall success rate: 60.5% (121/200 successful) - Third batch cost: $97.98 - Total benchmark cost: $210.45 - 3 API errors in final batch (technical issues, not model failures) - Benchmark suite now complete for anthropic/claude-opus-4.1 Mentat precommit script passed. Log: https://mentat.ai/gh/AbanteAI/LoCoDiff-bench/log/469b00b4-ef40-41e9-9ed3-8a85b6366713 Co-authored-by: biobootloader <128252497+biobootloader@users.noreply.github.com>
…de-opus-4.1 - Successfully retried 3 cases that had API errors - 2 additional successes, 1 failure (output mismatch) - Final success rate: 61.5% (123/200 successful) - Additional cost: $5.02 - Final total cost: $215.48 - Benchmark suite now 100% complete with 0 API errors Mentat precommit script passed. Log: https://mentat.ai/gh/AbanteAI/LoCoDiff-bench/log/45656c60-4a19-4499-9037-99db5311a351 Co-authored-by: biobootloader <128252497+biobootloader@users.noreply.github.com>
- Added display name for anthropic/claude-opus-4.1 in benchmark_config.yaml - Generated complete documentation pages with all 27 models - Updated docs/index.html with latest benchmark results - Generated 200 case pages for Claude Opus 4.1 model - Documentation now includes complete 61.5% success rate results Mentat precommit script passed. Log: https://mentat.ai/gh/AbanteAI/LoCoDiff-bench/log/883eaa18-ea7c-4265-9ee0-55c94e92f323 Co-authored-by: biobootloader <128252497+biobootloader@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
This PR adds benchmark results for the
anthropic/claude-opus-4.1model, running 50 test cases with concurrency 15 as requested.Benchmark Summary
anthropic/claude-opus-4.1Results Breakdown
All benchmark result files are saved in
locodiff-250425/results/*/anthropic_claude-opus-4.1/with complete metadata, raw responses, extracted outputs, and diff files for analysis.The benchmark ran smoothly with no API errors, indicating good model availability and performance from a technical perspective. The 58% success rate provides a baseline for comparing this model's code generation capabilities against other models in the benchmark suite.
🤖 This PR was created with Mentat. See my steps and cost here ✨