GPT-5.6 Terra Benchmarks, Specifications & Availability
Explore GPT-5.6 Terra from OpenAI: published specifications, source-linked vendor benchmarks, independent evaluator coverage and recorded pricing when available.
Compare GPT-5.6 Terra with other models →Explore data coverage
Published specifications
- Provider
- OpenAI
- Access
- Proprietary
- License
- Proprietary
- Context window
- 1.05M
- Total parameters
- Not published
- Active parameters
- Not published
- Released
- 2026-07-09
- Modalities
- file, image, text
- Family
- GPT-5.6
Model card · Announcement · Website · OpenRouter
Model notes
No GPT-6 Terra on OpenRouter as of 2026-09-23; GPT-5.6 Terra is the Terra tier in the Sol/Terra/Luna 5.6 family. Official benches TBD on refresh.
Pricing · OpenRouter
Dated cached OpenRouter rates in USD per 1M tokens. Open the dashboard for live enhancements. Per-metric endpoint minima can refer to different providers; they are not a guaranteed combined rate from one endpoint.
| Tier | Input / 1M | Output / 1M | Cached input / 1M | Cache write / 1M | Date & source |
|---|---|---|---|---|---|
| Default | [object Object] | [object Object] | [object Object] | [object Object] | 2026-10-03 · OpenRouter source |
Recorded pricing notes
min-healthy endpoint minima 2026-10-01; time-window overrides apply on a winning provider; min-healthy endpoint minima 2026-10-01; time-window overrides apply on a winning provider; min-healthy endpoint minima 2026-10-02; time-window overrides apply on a winning provider; min-healthy endpoint minima 2026-10-03; time-window overrides apply on a winning provider
Official / vendor benchmarks
Default headline records. Own-vendor, peer-vendor and third-party provenance remain visible in evidence; configurations may differ.
| Benchmark / evaluator | Headline score | Evidence |
|---|---|---|
| Agents' Last Exam | 50.4%unknown effort | All 1 recorded result & sources50.4% · raw 50.4 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| SWE-bench Pro | 63.4%unknown effort | All 1 recorded result & sources63.4% · raw 63.4 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| DeepSWE v1.1 | 69.6%unknown effort | All 1 recorded result & sources69.6% · raw 69.6 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| Terminal-Bench 2.1 | 87.4%unknown effort | All 1 recorded result & sources87.4% · raw 87.4 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| OSWorld 2.0 | 50.2%unknown effort | All 1 recorded result & sources50.2% · raw 50.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| BrowseComp | 87.5%unknown effort | All 1 recorded result & sources87.5% · raw 87.5 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| GPQA Diamond | 92.9%unknown effort | All 1 recorded result & sources92.9% · raw 92.9 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| MMMU Pro (no tools) | 80.7%unknown effort | All 1 recorded result & sources80.7% · raw 80.7 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 no tools https://openai.com/index/gpt-5-6 |
| MMMU Pro (with tools) | 82%unknown effort | All 1 recorded result & sources82% · raw 82 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 with tools https://openai.com/index/gpt-5-6 |
| FrontierMath Tier 4 | 68.3%unknown effort | All 1 recorded result & sources68.3% · raw 68.3 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 FrontierMath Tier 4 (v2) https://openai.com/index/gpt-5-6 |
| GDPval-AA v2 | 1593unknown effort | All 1 recorded result & sources1593 · raw 1593 Elo Headline · unknown effort · Own vendor Source/record date: 2026-07-09 Elo https://openai.com/index/gpt-5-6 |
| AA Intelligence Index (vendor-cited) | 55unknown effort | All 1 recorded result & sources55 · raw 55 index Headline · unknown effort · Own vendor Source/record date: 2026-07-09 Vendor-cited AA number in OpenAI table (still official announcement provenance) · Vendor-cited AA Intelligence Index v4.1 https://openai.com/index/gpt-5-6 |
| Management Consulting Tasks (Internal) | 37.2%unknown effort | All 1 recorded result & sources37.2% · raw 37.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 OpenAI internal https://openai.com/index/gpt-5-6 |
| Big Finance Bench | 51%unknown effort | All 1 recorded result & sources51% · raw 51 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| Artificial Analysis Coding Agent Index v1.1 | 77.4unknown effort | All 1 recorded result & sources77.4 · raw 77.4 index Headline · unknown effort · Own vendor Source/record date: 2026-07-09 Vendor-cited AA Coding Agent Index v1.1 https://openai.com/index/gpt-5-6 |
| GeneBench Pro | 23.3%unknown effort | All 1 recorded result & sources23.3% · raw 23.3 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| LifeSciBench | 56%unknown effort | All 1 recorded result & sources56% · raw 56 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| MedChemBench (Internal) | 35%unknown effort | All 1 recorded result & sources35% · raw 35 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 OpenAI internal https://openai.com/index/gpt-5-6 |
| HealthBench Professional | 57.7%unknown effort | All 1 recorded result & sources57.7% · raw 57.7 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| BenchCAD | 62.3%unknown effort | All 1 recorded result & sources62.3% · raw 62.3 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| BenchCAD (python tool) | 78.2%unknown effort | All 1 recorded result & sources78.2% · raw 78.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| Capture-the-Flag Challenges | 91.8%unknown effort | All 1 recorded result & sources91.8% · raw 91.8 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| SEC-Bench Pro | 57.7%unknown effort | All 1 recorded result & sources57.7% · raw 57.7 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| ExploitBench | 52.9%unknown effort | All 1 recorded result & sources52.9% · raw 52.9 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| ExploitGym | 23.2%unknown effort | All 1 recorded result & sources23.2% · raw 23.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| Internal Research Debugging Evaluation | 67.8%unknown effort | All 1 recorded result & sources67.8% · raw 67.8 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 OpenAI internal https://openai.com/index/gpt-5-6 |
| KernelGen 1P | 49.2%unknown effort | All 1 recorded result & sources49.2% · raw 49.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| NanoGPT | 14.5%unknown effort | All 1 recorded result & sources14.5% · raw 14.5 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| PostTrainBench Lite | 51.5%unknown effort | All 1 recorded result & sources51.5% · raw 51.5 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| RSI Index | 56.3%unknown effort | All 1 recorded result & sources56.3% · raw 56.3 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| GDP.PDF | 24.7%unknown effort | All 1 recorded result & sources24.7% · raw 24.7 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| FrontierMath Tier 1-3 (v2) | 84.9%unknown effort | All 1 recorded result & sources84.9% · raw 84.9 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 v2 https://openai.com/index/gpt-5-6 |
| AutomationBench | 15.2%unknown effort | All 1 recorded result & sources15.2% · raw 15.2 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 OpenAI GPT-5.6 card table (not necessarily v1.0.6) https://openai.com/index/gpt-5-6 |
| Toolathlon | 53.1%unknown effort | All 1 recorded result & sources53.1% · raw 53.1 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| MRCR 256K–512K | 89.6%unknown effort | All 1 recorded result & sources89.6% · raw 89.6 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 MRCR v2 8-needle https://openai.com/index/gpt-5-6 |
| MRCR 512K–1M | 72.5%unknown effort | All 1 recorded result & sources72.5% · raw 72.5 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 MRCR v2 8-needle https://openai.com/index/gpt-5-6 |
| GraphWalks BFS 256k f1 | 76.9 f1unknown effort | All 1 recorded result & sources76.9 f1 · raw 76.9 f1 Headline · unknown effort · Own vendor Source/record date: 2026-07-09 f1 https://openai.com/index/gpt-5-6 |
| GraphWalks BFS 1mil f1 | 71.2 f1unknown effort | All 1 recorded result & sources71.2 f1 · raw 71.2 f1 Headline · unknown effort · Own vendor Source/record date: 2026-07-09 f1 https://openai.com/index/gpt-5-6 |
| ARC-AGI-3 | 0.8%unknown effort | All 1 recorded result & sources0.8% · raw 0.8 % Headline · unknown effort · Own vendor Source/record date: 2026-07-09 https://openai.com/index/gpt-5-6 |
| HealthBench Hard | 32.7%unknown effort | All 1 recorded result & sources32.7% · raw 32.7 % Headline · unknown effort · Peer vendor Source/record date: 2026-09-22 length-adjusted (raw 34.3); as reported in GPT-6 Astra system card HealthBench table https://deploymentsafety.openai.com/gpt-6-astra |
| Terminal-Bench 4.0 | 21.5%max effort | All 1 recorded result & sources21.5% · raw 21.5 % Headline · max effort · Own vendor Source/record date: 2026-09-23 Official Terminal-Bench 4.0 board: GPT-5.6 Terra (max) + Codex; resolution rate 21.5% ±3.3 https://www.tbench.ai/leaderboard/terminal-bench/4.0 |
| FrontierCode 1.1 Main | 41.3%unknown effort | All 1 recorded result & sources41.3% · raw 41.3 % Headline · unknown effort · Own vendor Source/record date: 2026-08-01 Cognition FrontierCode 1.1 Main weighted score 41.3%; board export mirrored by evals.report (sourceUrl cognition.com) and matches llm-stats 0.413; Cognition changelog notes Terra/Luna pricing updates https://cognition.com/frontiercode |
Independent evaluators
Evaluator harnesses are distinct from vendor measurements. Missing coverage is not a failed test.
| Benchmark / evaluator | Headline score | Evidence |
|---|---|---|
| Intelligence Index · Artificial Analysis | 42max effort | All 6 recorded results & sources42 · raw 42 index Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra38 · raw 38 index Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-xhigh34 · raw 34 index Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-high30 · raw 30 index Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-medium27 · raw 27 index Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-low21 · raw 21 index Alternative · none effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-non-reasoning |
| Vals Index · Vals AI | 53.1%unknown effort | All 1 recorded result & sources53.1% · raw 53.09 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-03 Refreshed from current Vals leaderboard. Cost/test $5.78. https://www.vals.ai/benchmarks/vals_index |
| Bugs fixed /105 · Bug Hunt Bench | 32 fixesmax effort | All 4 recorded results & sources32 fixes · raw 32 fixes Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Harness: Codex CLI; effort max; 1 runs; evaluation 2026-08-27. Best documented score for this effort in Oct 1 README. Headline: best documented model run. https://github.com/phuryn/bug-hunt-bench20 fixes · raw 20 fixes Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Harness: Codex CLI; effort xhigh; 1 runs; evaluation 2026-08-28. Best documented score for this effort in Oct 1 README. https://github.com/phuryn/bug-hunt-bench18 fixes · raw 18 fixes Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Harness: Codex CLI; effort high; 1 runs; evaluation 2026-08-27. Best documented score for this effort in Oct 1 README. https://github.com/phuryn/bug-hunt-bench15 fixes · raw 15 fixes Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Harness: Codex CLI; effort medium; 1 runs; evaluation 2026-08-27. Best documented score for this effort in Oct 1 README. https://github.com/phuryn/bug-hunt-bench |
| ARC-AGI-1 · ARC Prize | 96.5%max effort | All 5 recorded results & sources96.5% · raw 96.5 % Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. Headline: best verified value; source matrix effort order breaks ties. https://arcprize.org/results/openai-gpt-5-6-terra94% · raw 94 % Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra92% · raw 92 % Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra77% · raw 77 % Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra60.2% · raw 60.2 % Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra |
| ARC-AGI-2 · ARC Prize | 83.9%max effort | All 5 recorded results & sources83.9% · raw 83.9 % Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. Headline: best verified value; source matrix effort order breaks ties. https://arcprize.org/results/openai-gpt-5-6-terra74.2% · raw 74.2 % Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra67.1% · raw 67.1 % Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra37.5% · raw 37.5 % Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra18.8% · raw 18.8 % Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra |
| ARC-AGI-3 · ARC Prize | 0.8%max effort | All 5 recorded results & sources0.8% · raw 0.8 % Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. Headline: best verified value; source matrix effort order breaks ties. https://arcprize.org/results/openai-gpt-5-6-terra0.7% · raw 0.65 % Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra0.5% · raw 0.49 % Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra0.1% · raw 0.08 % Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra0% · raw 0.01 % Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Semi-Private verified matrix. Published verified configuration. https://arcprize.org/results/openai-gpt-5-6-terra |
| GDPval-AA Elo · Artificial Analysis | 1432max effort | All 1 recorded result & sources1432 · raw 1432 Elo Headline · max effort · Independent evaluator Source/record date: 2026-09-23 GDPval-AA v2.1 Elo; GPT-5.6 Terra (max) https://artificialanalysis.ai/evaluations/gdpval-aa |
| Output speed · Artificial Analysis | 100 tok/smax effort | All 6 recorded results & sources100 tok/s · raw 100 tok/s Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra91 tok/s · raw 91 tok/s Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra-xhigh85 tok/s · raw 85 tok/s Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra-high84 tok/s · raw 84 tok/s Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra-medium80 tok/s · raw 80 tok/s Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra-low84 tok/s · raw 84 tok/s Alternative · none effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; Median output tokens/s; leaderboard rounds to whole tokens. https://artificialanalysis.ai/models/gpt-5-6-terra-non-reasoning |
| Cost per Intelligence Index task · Artificial Analysis | $1.40max effort | All 6 recorded results & sources$1.40 · raw 1.4 USD Headline · max effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra$0.63 · raw 0.63 USD Alternative · xhigh effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-xhigh$0.34 · raw 0.34 USD Alternative · high effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-high$0.18 · raw 0.18 USD Alternative · medium effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-medium$0.14 · raw 0.14 USD Alternative · low effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-low$0.14 · raw 0.14 USD Alternative · none effort · Independent evaluator Source/record date: 2026-10-03 Current AA model leaderboard; https://artificialanalysis.ai/models/gpt-5-6-terra-non-reasoning |
| Money gain · Andon Labs | $6843.21unknown effort | All 1 recorded result & sources$6843.21 · raw 6843.21 $ Headline · unknown effort · Independent evaluator Source/record date: 2026-10-03 Vending-Bench 2 net gain = final_value in the page public vb2 data module minus $500 starting balance. Full 66-model source checked. https://andonlabs.com/evals/vending-bench-2 |
| Blueprint Bench · Andon Labs | 30.8%unknown effort | All 1 recorded result & sources30.8% · raw 30.8 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-03 Blueprint-Bench 2 connectivity similarity; published fractional score multiplied by 100. https://andonlabs.com/evals/blueprint-bench-2 |
| Vibe Code Bench v1.1 · Vals AI | 74.6%unknown effort | All 1 recorded result & sources74.6% · raw 74.59 % Headline · unknown effort · Independent evaluator Source/record date: 2026-10-03 Refreshed from current Vals leaderboard. Harness: OpenHands. Cost/test $7.89. https://www.vals.ai/benchmarks/vibe-code |
Read how we select and source scores or the comparison guide.