Head-to-head comparison
GPT-5.6 Terra vs. GPT-5.4 mini
Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.
142
shared result series
21
distinct benchmarks
2026-08-16
latest retrieval
Change model selection
Specifications and cost
General specifications and standard API pricing
Undisclosed parameter counts and knowledge cutoffs stay marked as unavailable. The table shows direct standard token pricing, including documented context tiers. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.
| Attribute | GPT-5.6 Terra | GPT-5.4 mini |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Released | Jun 26, 2026 | Mar 2026 |
| Availability | Active | Active |
| Model type | Proprietary | Proprietary |
| Parameters | Not published | Not published |
| Architecture | Not published | Not published |
| Context window | 1,050,000 tokens | 400,000 tokens |
| Knowledge cutoff | February 16, 2026 | August 31, 2025 |
| Technical sources | Model, Context, Knowledge cutoff | Model |
| API pricing per 1M tokens | ||
| API input | $2 ≤272K / $4 >272K | $0.75 |
| API output | $12 ≤272K / $18 >272K | $4.5 |
| Cache read | $0.2 ≤272K / $0.4 >272K | $0.07 |
| Cache write (5 min.) | $2.5 ≤272K / $5 >272K | Not listed |
| Cache write (1 hr.) | Not listed | Not listed |
| Price verified | 08/11/2026Source | 07/30/2026Source |
GPT-5.6 TerraGPT-5.4 mini
API inputUSD per 1M tokens, base tier
GPT-5.6 Terra$2
GPT-5.4 mini$0.75Lower price
API outputUSD per 1M tokens, base tier
GPT-5.6 Terra$12
GPT-5.4 mini$4.5Lower price
Cache readUSD per 1M tokens, base tier
GPT-5.6 Terra$0.2
GPT-5.4 mini$0.075Lower price
Performance
Shared benchmarks
Only published results with the same benchmark version, task, metric, and comparison cohort are paired. Different reasoning levels, agents, harnesses, or output limits appear directly in the table.
142 shared result series
GPT-5.6 TerraGPT-5.4 mini
Arena Text, Style ControlOverall. Higher is better. Test setup not documented as identical
GPT-5.6 Terra1,464.935 arena pointsHigher measured value
GPT-5.4 mini1,448.205 arena points
Code MigrationOverall. Higher is better. No documented setup difference
GPT-5.6 Terra36.445 %Winner
GPT-5.4 mini12.936 %
CorpFin v2Overall. Higher is better. No documented setup difference
GPT-5.6 Terra65.307 %Winner
GPT-5.4 mini60.917 %
EMBOverall. Higher is better. No documented setup difference
GPT-5.6 Terra67.614 %Winner
GPT-5.4 mini45.426 %
Finance Agent (v2)Overall. Higher is better. No documented setup difference
GPT-5.6 Terra52.42 %Winner
GPT-5.4 mini45.36 %
GPQA DiamondOverall. Higher is better. No documented setup difference
GPT-5.6 Terra90.909 %Winner
GPT-5.4 mini83.08 %
Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference
GPT-5.6 Terra0.417 %Winner
GPT-5.4 mini0 %
IOIOverall. Higher is better. No documented setup difference
GPT-5.6 Terra65.25 %Winner
GPT-5.4 mini6.417 %
Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference
GPT-5.6 Terra40.865 %Winner
GPT-5.4 mini12.5 %
LiveCodeBench v6Overall. Higher is better. No documented setup difference
GPT-5.6 Terra85.93 %Winner
GPT-5.4 mini81.465 %
MMLU-ProOverall. Higher is better. No documented setup difference
GPT-5.6 Terra86.659 %Winner
GPT-5.4 mini84.554 %
MMMU-ProOverall. Higher is better. No documented setup difference
GPT-5.6 Terra86.474 %Winner
GPT-5.4 mini79.249 %
| Benchmark and task | GPT-5.6 Terra | GPT-5.4 mini | Comparability | Source |
|---|---|---|---|---|
| Arena Text, Style ControlOverall | 1,464.935 arena pointsHigher measured value 95% confidence interval: ±5.686The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,448.205 arena points 95% confidence interval: ±3.859The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:overallTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Snapshot: gpt-5.6-terra-xhigh, Reasoning: xhigh, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlGPT-5.6 Terra: Style-controlled Arena rating from 16,008 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 59,631 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationOverall | 36.445 %Winner | 12.936 % | 1accuracyHigher is betterCohort: vals-code-migration:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-code-migration-2026-08-16-107652dc6bfcRetrieved 2026-08-16 |
| CorpFin v2Overall | 65.307 %Winner | 60.917 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-corp_fin_v2-2026-08-16-45bbf84f57abRetrieved 2026-08-16 |
| EMBOverall | 67.614 %Winner | 45.426 % | 1accuracyHigher is betterCohort: vals-emb:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-emb-2026-08-16-68d5d1d47a97Retrieved 2026-08-16 |
| Finance Agent (v2)Overall | 52.42 %Winner | 45.36 % | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-fabv2-2026-08-16-8853da43085dRetrieved 2026-08-16 |
| GPQA DiamondOverall | 90.909 %Winner | 83.08 % | 1accuracyHigher is betterCohort: vals-gpqa:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-gpqa-2026-08-16-7f8363eaa2dbRetrieved 2026-08-16 |
| Harvey's Legal Agent BenchmarkOverall · Task fully resolved | 0.417 %Winner | 0 % | 1task resolution rateHigher is betterCohort: vals-legal-agent-benchmark:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-hlab-2026-08-16-bae90566df2fRetrieved 2026-08-16 |
| IOIOverall | 65.25 %Winner | 6.417 % | 1accuracyHigher is betterCohort: vals-ioi:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-ioi-2026-08-16-d9164ef483aaRetrieved 2026-08-16 |
| Legal Research BenchOverall · All-pass | 40.865 %Winner | 12.5 % | 1all-pass rateHigher is betterCohort: vals-legal-research:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-legal_research-2026-08-16-af296c00ecaaRetrieved 2026-08-16 |
| LiveCodeBench v6Overall | 85.93 %Winner | 81.465 % | 1accuracyHigher is betterCohort: vals-lcb:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-lcb-2026-08-16-17b9c147080dRetrieved 2026-08-16 |
| MMLU-ProOverall | 86.659 %Winner | 84.554 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-mmlu_pro-2026-08-16-e14d5bb3a34cRetrieved 2026-08-16 |
| MMMU-ProOverall | 86.474 %Winner | 79.249 % | 1accuracyHigher is betterCohort: vals-mmmu:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmmu-2026-08-16-37c9a8c7e21fRetrieved 2026-08-16 |
| MortgageTaxOverall | 67.329 %Winner | 63.514 % | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-mortgage_tax-2026-08-16-e78c985dbb1cRetrieved 2026-08-16 |
| ProgramBenchOverall | 0.5 %Winner | 0 % | 1accuracyHigher is betterCohort: vals-programbench:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-programbench-2026-08-16-08d9727662ddRetrieved 2026-08-16 |
| SAGEOverall | 47.003 % | 50.813 %Winner | 1accuracyHigher is betterCohort: vals-sage:1:overallNo documented setup differenceWinner: GPT-5.4 miniGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-sage-2026-08-16-a416dd25affbRetrieved 2026-08-16 |
| SWE-bench VerifiedOverall | 75.2 %Winner | 73 % | 1accuracyHigher is betterCohort: vals-swebench:1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-swebench-2026-08-16-c133071b8167Retrieved 2026-08-16 |
| TaxEval v2Overall | 76.166 %Winner | 71.218 % | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-12Import ID: vals-tax_eval_v2-2026-08-16-5bbd495dac53Retrieved 2026-08-16 |
| Terminal-Bench 2.1Overall | 73.408 %Winner | 54.682 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-terminal-bench-2-1-2026-08-16-b3f38c18bea4Retrieved 2026-08-16 |
| Vals IndexOverall | 56.535 %Winner | 39.632 % | 2weighted index scoreHigher is betterCohort: vals-vals-index:2:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-14Import ID: vals-vals_index-2026-08-16-ab641d89d6edRetrieved 2026-08-16 |
| Vals Multimodal IndexOverall | 65.07 %Winner | 54.219 % | 1.2weighted index scoreHigher is betterCohort: vals-vals-multimodal-index:1.2:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: No further settings disclosed GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17 | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-11Import ID: vals-vals_multimodal_index-2026-08-16-9f3ebb221408Retrieved 2026-08-16 |
| Vibe Code Bench v1.1Overall | 67.842 %Winner | 47.969 % | 1.1accuracyHigher is betterCohort: vals-vibe-code-bench:1.1:overallNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Harness: OpenHands GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Harness: OpenHands | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-13Import ID: vals-vibe-code-2026-08-16-a73c9770bcc4Retrieved 2026-08-16 |
| Arena Text, Style ControlBusiness, management and finance | 1,468.921 arena pointsHigher measured value 95% confidence interval: ±11.251The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,460.132 arena points 95% confidence interval: ±6.68The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:industry_business_and_management_and_financial_operationsTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Snapshot: gpt-5.6-terra-xhigh, Reasoning: xhigh, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlGPT-5.6 Terra: Style-controlled Arena rating from 3,050 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 12,114 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCLIReasoning: xhigh | 26.645 %Winner | 8.2 % | 1accuracyHigher is betterCohort: vals-code-migration:1:cliNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Exact PagesReasoning: xhigh | 63.17 %Winner | 61.538 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:exact_pagesNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| EMBCompsReasoning: xhigh | 72.539 %Winner | 40.488 % | 1accuracyHigher is betterCohort: vals-emb:1:compsNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
| Finance Agent (v2)AdjustmentsReasoning: xhigh | 50.442 %Winner | 45.366 % | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:adjustmentsNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-fabv2-2026-08-11-3e5c93a1857cRetrieved 2026-08-11 |
| GPQA DiamondFew-Shot CoTReasoning: xhigh | 89.899 %Winner | 84.848 % | 1accuracyHigher is betterCohort: vals-gpqa:1:diamond_few_shot_cotNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-gpqa-2026-08-11-d3325290a404Retrieved 2026-08-11 |
| IOIIOI 2024Reasoning: xhigh | 66.167 %Winner | 6.667 % | 1accuracyHigher is betterCohort: vals-ioi:1:ioi2024No documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-ioi-2026-08-11-2dfe8b4b8b7bRetrieved 2026-08-11 |
| Legal Research BenchAdministrative / Regulatory · All-passReasoning: xhigh | 57.576 %Winner | 15.152 % | 1all-pass rateHigher is betterCohort: vals-legal-research:1:administrative_regulatoryNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-legal_research-2026-08-11-326cf7aaca58Retrieved 2026-08-11 |
| LiveCodeBench v6EasyReasoning: xhigh | 97.205 % | 97.826 %Winner | 1accuracyHigher is betterCohort: vals-lcb:1:easyNo documented setup differenceWinner: GPT-5.4 miniGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-lcb-2026-08-11-0342bcc364e4Retrieved 2026-08-11 |
| MMLU-ProBiologyReasoning: xhigh | 93.026 %Winner | 91.353 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:biologyNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmlu_pro-2026-08-11-b8e53828d746Retrieved 2026-08-11 |
| MortgageTaxNumerical ExtractionReasoning: xhigh | 74.324 % | 79.412 %Winner | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:numerical_extractionNo documented setup differenceWinner: GPT-5.4 miniGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mortgage_tax-2026-08-11-5ea6eef3e482Retrieved 2026-08-11 |
| ProgramBenchAlmost ResolvedReasoning: xhigh | 6.5 %Winner | 3 % | 1accuracyHigher is betterCohort: vals-programbench:1:almostNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-programbench-2026-08-11-b6f8959fbdfeRetrieved 2026-08-11 |
| SAGEAutomataReasoning: xhigh | 38.889 % | 41.27 %Winner | 1accuracyHigher is betterCohort: vals-sage:1:automataNo documented setup differenceWinner: GPT-5.4 miniGPT-5.6 Terra: Reasoning: xhigh, Output limit: 30,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 30,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-sage-2026-08-11-f1f79f6d8af3Retrieved 2026-08-11 |
| SWE-bench Verified<15 min fixReasoning: xhigh | 77.835 % | 85.567 %Winner | 1accuracyHigher is betterCohort: vals-swebench:1:<15 min fixNo documented setup differenceWinner: GPT-5.4 miniGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-08Import ID: vals-swebench-2026-08-11-a6212cd857d8Retrieved 2026-08-11 |
| TaxEval v2CorrectnessReasoning: xhigh | 65.822 %Winner | 63.042 % | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:correctnessNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-tax_eval_v2-2026-08-11-d05a40bf446fRetrieved 2026-08-11 |
| Terminal-Bench 2.1Easy | 83.333 %Higher measured value | 75 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:easyTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: high, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-terminal-bench-2-1-2026-08-11-8f0ebd72a609Retrieved 2026-08-11 |
| Vals IndexCorpFin v2Reasoning: xhigh | 66.434 %Winner | 60.256 % | 1.2benchmark scoreHigher is betterCohort: vals-vals-index:1.2:corp_fin_v2No documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_index-2026-08-11-e79d08d59249Retrieved 2026-08-11 |
| Vals Multimodal IndexCorpFin v2Reasoning: xhigh | 66.434 %Winner | 60.256 % | 1.2benchmark scoreHigher is betterCohort: vals-vals-multimodal-index:1.2:corp_fin_v2No documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_multimodal_index-2026-08-11-fa704862df19Retrieved 2026-08-11 |
| Arena Text, Style ControlChinese | 1,517.892 arena pointsHigher measured value 95% confidence interval: ±18.554The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,478.62 arena points 95% confidence interval: ±11.489The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:chineseTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Snapshot: gpt-5.6-terra-xhigh, Reasoning: xhigh, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlGPT-5.6 Terra: Style-controlled Arena rating from 1,102 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 3,250 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCOBOLReasoning: xhigh | 65.844 %Winner | 27.143 % | 1accuracyHigher is betterCohort: vals-code-migration:1:cobolNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Max Fitting ContextReasoning: xhigh | 66.317 %Winner | 60.956 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:max_fitting_contextNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| EMBDataroomReasoning: xhigh | 81.591 %Winner | 48.286 % | 1accuracyHigher is betterCohort: vals-emb:1:dataroomNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
| Finance Agent (v2)All-PassReasoning: xhigh | 40.448 %Winner | 32.369 % | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:all_passNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-fabv2-2026-08-11-3e5c93a1857cRetrieved 2026-08-11 |
| GPQA DiamondZero-Shot CoTReasoning: xhigh | 91.919 %Winner | 81.313 % | 1accuracyHigher is betterCohort: vals-gpqa:1:diamond_zero_shot_cotNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-gpqa-2026-08-11-d3325290a404Retrieved 2026-08-11 |
| IOIIOI 2025Reasoning: xhigh | 64.333 %Winner | 6.167 % | 1accuracyHigher is betterCohort: vals-ioi:1:ioi2025No documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-ioi-2026-08-11-2dfe8b4b8b7bRetrieved 2026-08-11 |
| Legal Research BenchAdministrative / Regulatory · Weighted partial creditReasoning: xhigh | 86.696 %Winner | 69.72 % | 1weighted partial-credit scoreHigher is betterCohort: vals-legal-research:1:weighted_pass__administrative_regulatoryNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-legal_research-2026-08-11-326cf7aaca58Retrieved 2026-08-11 |
| LiveCodeBench v6HardReasoning: xhigh | 72.857 %Winner | 61.714 % | 1accuracyHigher is betterCohort: vals-lcb:1:hardNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-lcb-2026-08-11-0342bcc364e4Retrieved 2026-08-11 |
| MMLU-ProBusinessReasoning: xhigh | 90.875 %Winner | 89.734 % | 1accuracyHigher is betterCohort: vals-mmlu-pro:1:businessNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mmlu_pro-2026-08-11-b8e53828d746Retrieved 2026-08-11 |
| MortgageTaxSemantic ExtractionReasoning: xhigh | 60.334 %Winner | 47.615 % | 1accuracyHigher is betterCohort: vals-mortgage-tax:1:semantic_extractionNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-mortgage_tax-2026-08-11-5ea6eef3e482Retrieved 2026-08-11 |
| ProgramBenchFully ResolvedReasoning: xhigh | 0.5 %Winner | 0 % | 1accuracyHigher is betterCohort: vals-programbench:1:strictNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-programbench-2026-08-11-b6f8959fbdfeRetrieved 2026-08-11 |
| SAGECalculusReasoning: xhigh | 59.259 %Tie | 59.259 %Tie | 1accuracyHigher is betterCohort: vals-sage:1:calculusNo documented setup differenceTieGPT-5.6 Terra: Reasoning: xhigh, Output limit: 30,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 30,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-sage-2026-08-11-f1f79f6d8af3Retrieved 2026-08-11 |
| SWE-bench Verified>4 hoursReasoning: xhigh | 33.333 %Winner | 0 % | 1accuracyHigher is betterCohort: vals-swebench:1:>4 hoursNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-08Import ID: vals-swebench-2026-08-11-a6212cd857d8Retrieved 2026-08-11 |
| TaxEval v2Stepwise ReasoningReasoning: xhigh | 86.509 %Winner | 79.395 % | 2accuracyHigher is betterCohort: vals-tax-eval-v2:2:stepwise_reasoningNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-tax_eval_v2-2026-08-11-d05a40bf446fRetrieved 2026-08-11 |
| Terminal-Bench 2.1Hard | 63.333 %Higher measured value | 34.444 % | 2.1accuracyHigher is betterCohort: vals-terminal-bench-2-1:2.1:hardTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: high, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-terminal-bench-2-1-2026-08-11-8f0ebd72a609Retrieved 2026-08-11 |
| Vals IndexFinance Agent v2Reasoning: xhigh | 52.626 %Winner | 45.407 % | 1.2benchmark scoreHigher is betterCohort: vals-vals-index:1.2:finance_agentNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_index-2026-08-11-e79d08d59249Retrieved 2026-08-11 |
| Vals Multimodal IndexFinance Agent v2Reasoning: xhigh | 52.626 %Winner | 45.407 % | 1.2benchmark scoreHigher is betterCohort: vals-vals-multimodal-index:1.2:finance_agentNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-vals_multimodal_index-2026-08-11-fa704862df19Retrieved 2026-08-11 |
| Arena Text, Style ControlCoding | 1,515.671 arena pointsHigher measured value 95% confidence interval: ±9.356The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | 1,496.777 arena points 95% confidence interval: ±6.045The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result. | style-control-v1style-controlled Bradley-Terry ratingHigher is betterCohort: lmarena-text-style-control:2026-08-12:codingTest setup not documented as identicalNo direct winnerGPT-5.6 Terra: Snapshot: gpt-5.6-terra-xhigh, Reasoning: xhigh, Harness: Arena text, style control GPT-5.4 mini: Snapshot: gpt-5.4-mini-high, Reasoning: high, Harness: Arena text, style controlGPT-5.6 Terra: Style-controlled Arena rating from 4,522 anonymous pairwise votes.GPT-5.4 mini: Style-controlled Arena rating from 16,275 anonymous pairwise votes. | ArenaType: Official benchmark leaderboardLeaderboard updated: 2026-08-12Import ID: lmarena-eeedf9209373a7f9b324bfb839e7b02fedacd1e5-1611ed64f3f1Retrieved 2026-08-16 |
| Code MigrationCode QualityReasoning: xhigh | 3.305 / 5Winner | 3.053 / 5 | 1mean reviewer code-quality ratingHigher is betterCohort: vals-code-migration:1:code_qualityNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-code-migration-2026-08-11-3a2d5ce718a3Retrieved 2026-08-11 |
| CorpFin v2Shared Max ContextReasoning: xhigh | 66.434 %Winner | 60.256 % | 2accuracyHigher is betterCohort: vals-corp-fin-v2:2:shared_max_contextNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-corp_fin_v2-2026-08-11-d9e58d191cd9Retrieved 2026-08-11 |
| EMBDCFReasoning: xhigh | 63.137 %Winner | 40.605 % | 1accuracyHigher is betterCohort: vals-emb:1:dcfNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-emb-2026-08-11-e3c855e236e4Retrieved 2026-08-11 |
| Finance Agent (v2)ComparablesReasoning: xhigh | 45.218 %Winner | 40.273 % | 2accuracyHigher is betterCohort: vals-finance-agent-v2:2:comparablesNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-fabv2-2026-08-11-3e5c93a1857cRetrieved 2026-08-11 |
| Legal Research BenchBusiness & Commercial · All-passReasoning: xhigh | 37.209 %Winner | 11.628 % | 1all-pass rateHigher is betterCohort: vals-legal-research:1:business_commercialNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-06Import ID: vals-legal_research-2026-08-11-326cf7aaca58Retrieved 2026-08-11 |
| LiveCodeBench v6MediumReasoning: xhigh | 87.728 %Winner | 84.856 % | 1accuracyHigher is betterCohort: vals-lcb:1:mediumNo documented setup differenceWinner: GPT-5.6 TerraGPT-5.6 Terra: Reasoning: xhigh, Output limit: 128,000 tokens GPT-5.4 mini: Snapshot: gpt-5.4-mini-2026-03-17, Reasoning: xhigh, Output limit: 128,000 tokens | Vals AIType: Independent evaluationLeaderboard updated: 2026-08-09Import ID: vals-lcb-2026-08-11-0342bcc364e4Retrieved 2026-08-11 |
The table shows 64 diversified result series from 142 shared results. This prevents one benchmark with many subtasks from crowding out every other test.
How to read this comparison
Winning one benchmark is not an overall verdict
Your workload, budget, and required context length matter most. A coding benchmark says little about visual understanding or agent performance.
Prices use official API rates per 1M tokens. Web subscriptions and cloud platform surcharges can differ.
Every result links to its measurement source and retrieval date. Labels distinguish vendor reports, official benchmark leaderboards, and independent evaluations. Disputed or archived results are excluded.
More matchups
Related LLM comparisons
Compare GPT-5.6 Terra and GPT-5.4 mini with other leading models using the same data and benchmark logic.
View the complete LLM comparison