bix-41-q5 | Current BixBench results

New token-optimization experiment

Three additional GPT-5.6 Sol Galaxy replicates are available separately. Previous results remain unchanged.

View the October 2026 batch and its run records

Question

In a bacterial swarming assay compared mixed cultures of Strain 287 and Strain 98 at various ratios, which ratio produced colonies with a mean area and circularity value that is most similar to the mean area and circularity values observed in Strain 1?

Accepted answer: 5:1
Domain: general

Comparison

Pass counts and token totals across the same three replicates.

Agent	Galaxy-API code	Open-ended code
Codex GPT-5.5	3/3 replicate 1 replicate 2 replicate 3 2.08M total	3/3 replicate 1 replicate 2 replicate 3 246.3K total
Codex GPT-5.6 Sol	3/3 replicate 1 replicate 2 replicate 3 4.68M total	3/3 replicate 1 replicate 2 replicate 3 263.6K total
Codex GPT-5.6 Luna	3/3 replicate 1 replicate 2 replicate 3 22.49M total	3/3 replicate 1 replicate 2 replicate 3 303.9K total
DeepSeek V4 Pro via Codex	3/3 replicate 1 replicate 2 replicate 3 10.45M total	3/3 replicate 1 replicate 2 replicate 3 394.1K total
DeepSeek V4 Pro via Claude Codesuperseded	3/3 replicate 1 replicate 2 replicate 3 8.84M total	3/3 replicate 1 replicate 2 replicate 3 410.0K total

Execution condition

Galaxy-API code with skills

Codex GPT-5.5

3 primary replicates

Codex GPT-5.5

replicate 1

passUDT-only

Answer5:1

Workflow

Galaxy UDTcolony morphology summary

Total tokens: 602.0K
Output: 6.1K

Galaxy history Run trace

Codex GPT-5.5

replicate 2

passUDT-only

Answer5:1

Workflow

Galaxy UDTcolony morphology summary

Total tokens: 597.2K
Output: 5.6K

Galaxy history Run trace

Codex GPT-5.5

replicate 3

passUDT-only

Answer5:1

Workflow

Galaxy UDTcolony morphology summarymean summarydistance comparison

Total tokens: 882.1K
Output: 10.2K

Galaxy history Run trace

Codex GPT-5.6 Sol

3 primary replicates

Codex GPT-5.6 Sol

replicate 1

pass

Answer5:1

Workflow

Csv To TabularDatamash summary

Total tokens: 935.7K
Output: 6.0K

Galaxy history Run trace

Codex GPT-5.6 Sol

replicate 2

pass

Answer5:1

Workflow

Csv To TabularFilter TabularMatch rowsGroup and aggregateAdd derived column

Total tokens: 2.09M
Output: 8.9K

Galaxy history Run trace

Codex GPT-5.6 Sol

replicate 3

passUDT + standard tools

Answer5:1

Workflow

Csv To TabularDatamash summaryAdd derived columnGalaxy UDTcolony morphology summary

Total tokens: 1.66M
Output: 12.2K

Galaxy history Run trace

Codex GPT-5.6 Luna

3 primary replicates

Codex GPT-5.6 Luna

replicate 1

pass

Answer5:1

Workflow

Csv To TabularFilter TabularDatamash summaryAdd derived columncolony morphology summary

Total tokens: 4.57M
Output: 19.9K

Galaxy history Run trace

Codex GPT-5.6 Luna

replicate 2

passUDT + standard tools

Answer5:1

Workflow

Csv To TabularDatamash summaryGroup and aggregateGalaxy UDTcolony morphology summarydistance comparison

Total tokens: 4.09M
Output: 22.4K

Galaxy history Run trace

Codex GPT-5.6 Luna

replicate 3

pass

Answer5:1

Workflow

Csv To TabularDatamash summaryGroup and aggregatemore stepscolony morphology summaryscikit-learn

Total tokens: 13.83M
Output: 49.0K

Galaxy history Run trace

DeepSeek V4 Pro via Codex

3 primary replicates

DeepSeek V4 Pro via Codex

replicate 1

pass

Answer5:1

Workflow

Datamash summaryCsv To TabularAdd derived columnFilter TabularTp Sort Header Toolcolony morphology summary

Total tokens: 4.81M
Output: 27.2K

Galaxy history Run trace

DeepSeek V4 Pro via Codex

replicate 2

pass

Answer5:1

Workflow

Filter TabularGroup and aggregateAdd derived columncolony morphology summarymean summary

Total tokens: 2.08M
Output: 17.4K

Galaxy history Run trace

DeepSeek V4 Pro via Codex

replicate 3

pass

Answer5:1

Workflow

Csv To TabularFilter TabularGroup and aggregateAdd derived columnTp Sort Header Tool

Total tokens: 3.57M
Output: 21.2K

Galaxy history Run trace

DeepSeek V4 Pro via Claude CodeSuperseded

3 primary replicates

DeepSeek V4 Pro via Claude Code

replicate 1

passUDT + standard tools

Answer5:1

Workflow

Upload datasetDatamash summaryGalaxy UDTcolony morphology summarydistance comparison

Total tokens: 2.66M
Output: 24.8K

Galaxy history Run trace

DeepSeek V4 Pro via Claude Code

replicate 2

passUDT + standard tools

Answer5:1

Workflow

Datamash summaryGalaxy UDTcolony morphology summary

Total tokens: 1.70M
Output: 18.5K

Galaxy history Run trace

DeepSeek V4 Pro via Claude Code

replicate 3

passUDT + standard tools

Answer5:1

Workflow

Filter Tabularmean summarycolony morphology summarydistance comparison

Total tokens: 4.48M
Output: 38.1K

Galaxy history Run trace

Execution condition

Open-ended code with skills

Codex GPT-5.5

3 primary replicates

Codex GPT-5.5

replicate 1

pass

Answer5:1

Workflow

Local analysismean summarycolony morphology summary

Total tokens: 90.4K
Output: 1.4K

Run trace

Codex GPT-5.5

replicate 2

pass

Answer5:1

Workflow

Local analysiscolony morphology summarymean summarydistance comparison

Total tokens: 105.8K
Output: 1.7K

Run trace

Codex GPT-5.5

replicate 3

pass

Answer5:1

Workflow

Local analysis

Total tokens: 50.1K
Output: 803

Run trace

Codex GPT-5.6 Sol

3 primary replicates

Codex GPT-5.6 Sol

replicate 1

pass

Answer5:1

Workflow

Local analysis

Total tokens: 93.1K
Output: 837

Run trace

Codex GPT-5.6 Sol

replicate 2

pass

Answer5:1

Workflow

Local analysis

Total tokens: 76.8K
Output: 948

Run trace

Codex GPT-5.6 Sol

replicate 3

pass

Answer5:1

Workflow

Local analysis

Total tokens: 93.7K
Output: 1.1K

Run trace

Codex GPT-5.6 Luna

3 primary replicates

Codex GPT-5.6 Luna

replicate 1

pass

Answer5:1

Workflow

Local analysis

Total tokens: 91.2K
Output: 2.3K

Run trace

Codex GPT-5.6 Luna

replicate 2

pass

Answer5:1

Workflow

Local analysis

Total tokens: 111.3K
Output: 2.5K

Run trace

Codex GPT-5.6 Luna

replicate 3

pass

Answer5:1

Workflow

Local analysis

Total tokens: 101.5K
Output: 1.6K

Run trace

DeepSeek V4 Pro via Codex

3 primary replicates

DeepSeek V4 Pro via Codex

replicate 1

pass

Answer5:1

Workflow

Local analysiscolony morphology summarymean summarySciPyscikit-learn

Total tokens: 112.7K
Output: 5.0K

Run trace

DeepSeek V4 Pro via Codex

replicate 2

pass

Answer5:1

Workflow

Local analysiscolony morphology summarymean summarySciPydistance comparison

Total tokens: 102.9K
Output: 3.2K

Run trace

DeepSeek V4 Pro via Codex

replicate 3

pass

Answer5:1

Workflow

Local analysiscolony morphology summarymean summarydistance comparison

Total tokens: 178.5K
Output: 5.6K

Run trace

DeepSeek V4 Pro via Claude CodeSuperseded

3 primary replicates

DeepSeek V4 Pro via Claude Code

replicate 1

pass

Answer5:1

Workflow

Local analysiscolony morphology summarydistance comparison

Total tokens: 99.3K
Output: 1.9K

Run trace

DeepSeek V4 Pro via Claude Code

replicate 2

pass

Answer5:1

Workflow

Local analysiscolony morphology summarymean summarydistance comparison

Total tokens: 179.9K
Output: 2.9K

Run trace

DeepSeek V4 Pro via Claude Code

replicate 3

pass

Answer5:1

Workflow

Local analysiscolony morphology summarydistance comparison

Total tokens: 130.8K
Output: 3.4K

Run trace

bix-41-q5 · October 2026 · three additional replicates

GPT-5.6 Sol · token optimization

Galaxy-API code, high reasoning, Fast service. Previous runs are retained above.

Galaxy-API code · October 2026

replicate 1

Correct

Answer5:1

Total tokens: 848.2K
Input: 838.4K
Cached input: 796.2K
Output: 9.8K

Run trace Final answer Galaxy history

Galaxy-API code · October 2026

replicate 2

Correct

Answer5:1

Total tokens: 845.8K
Input: 840.2K
Cached input: 765.4K
Output: 5.6K

Run trace Final answer Galaxy history

Galaxy-API code · October 2026

replicate 3

Correct

Answer5:1

Total tokens: 608.0K
Input: 603.0K
Cached input: 538.0K
Output: 5.0K

Run trace Final answer Galaxy history

Open-ended code averaged 0.09M tokens; previous website Galaxy averaged 1.56M; the new Galaxy runs averaged 0.77M, or 8.73× Open-ended code. Each mean covers three replicates of this same task with GPT-5.6 Sol.

Cached input is already included in input tokens. Full nested tool/job evidence is in each HF run's galaxy_execution_records.tar.gz, with an accompanying member index.

Optimization overview and all 150 runs