OpenAIReasoning modelGPT-5.6 Terra
GPT-5.6 Terra is an OpenAI proprietary reasoning model, released 9 Jul 2026. Its best configuration (max reasoning effort) ranks #62 of 751 on the BenchLeader Index at 63.7 ±4.9, in the upper half. The ± is the point: 141 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in reasoning (71) and lowest in knowledge (57). At $4.50 per million tokens blended it is among the most expensive ranked models. Output speed of 100 tokens per second puts it faster than most, with a first answer in 176.8 s. It has been measured at 7 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 29 Sept 2026.
- Blended price
- $4.50/M
- $2.00 in · $12.00 out
- Output speed
- 100 tok/s
- measured by Artificial Analysis
- First answer
- 177 s
- first token 2.51 s
- Context
- 1.1M
- Full answer
- 182 s
- median, reasoning included
- Cost per run
- $1.40
- one full Intelligence Index run
- Released
- 9 Jul 2026
- Overall index64
- Reasoning71
- Coding65
- Agents & tools59
- Maths70
- Knowledge57
- Instruction following70
- Multimodal64
- Long context67
- Composite68
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 49.8 | #364 | 84 tok/s | 0.81 s | $0.0044 | Agents & tools 44 · Coding 48 · Composite 53 · Knowledge 51 · Long context 54 · Maths 47 · Multimodal 50 · Reasoning 51 |
| low | 56.7 | #179 | 80 tok/s | 1.77 s | $0.0044 | Agents & tools 55 · Coding 54 · Composite 61 · Instruction following 60 · Knowledge 59 · Long context 61 · Maths 63 · Multimodal 59 · Reasoning 52 |
| medium | 57.2 | #173 | 84 tok/s | 2.42 s | $0.0044 | Agents & tools 54 · Coding 55 · Composite 64 · Instruction following 62 · Knowledge 60 · Long context 62 · Multimodal 60 · Reasoning 56 |
| high | 61.5 | #93 | 85 tok/s | 3.47 s | $0.0044 | Agents & tools 66 · Coding 63 · Composite 69 · Instruction following 64 · Knowledge 60 · Long context 64 · Multimodal 62 · Reasoning 65 |
| xhigh | 63.0 | #70 | 91 tok/s | 29 s | $0.0044 | Agents & tools 63 · Coding 60 · Composite 73 · Human preference 66 · Instruction following 65 · Knowledge 60 · Long context 65 · Maths 67 · Multimodal 62 · Reasoning 64 |
| maxbest | 63.7 | #62 | 100 tok/s | 177 s | $0.0044 | Agents & tools 59 · Coding 65 · Composite 68 · Instruction following 70 · Knowledge 57 · Long context 67 · Maths 70 · Multimodal 64 · Reasoning 71 |
| not stated | – | – | 65 tok/s | 2.51 s | $0.0044 | Coding 55 · Human preference 55 · Reasoning 49 |
What more thinking costs
Turning the effort up buys index points and multiplies the bill. Here max costs 10× no reasoning for +13.9 on the index — 4.2 points per doubling of spend. Compare that with other models.
How its index has moved
8 Sept 2026 to 3 Oct 2026- max
- xhigh
- high
- medium
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| GPQA Diamond | 77.3%#130 | 87.4%#67 | – | – | – | 93.3%#17 | – | Epoch AI Benchmarking Hub | |
| SimpleBench | – | – | – | – | 48.9%#45 | – | 48.9%#45 | SimpleBench | |
| LMArena Hard Prompts | – | – | – | – | 1489#48 | – | – | LMArena | |
| LiveBench Reasoningnot in index | – | – | – | – | – | 90.6%#11 | – | LiveBench | |
| GPQA Diamond (AA)not in index | 74.7%#252 | 84.3%#140 | 87.2%#98 | 89.6%#70 | 90.8%#55 | 92.5%#35 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 11.4%#273 | 29.2%#141 | 33.3%#123 | 38.5%#90 | 41.9%#68 | 42.9%#55 | – | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | – | – | – | – | 90.9%#24 | – | – | Vals AI | |
| Kagi LLM Benchmark | – | – | – | – | – | – | 51.3%#82 | Kagi LLM Benchmark | |
| ARC-AGI-1 | – | 60.2%#135 | 77.0%#107 | 92.0%#53 | 94.0%#40 | 96.5%#21 | – | ARC Prize | |
| ARC-AGI-2 | – | 18.8%#122 | 37.5%#105 | 67.1%#60 | 74.2%#46 | 83.9%#33 | – | ARC Prize | |
| ARC-AGI-3 | – | 0.0%#51 | 0.1%#48 | 0.5%#31 | 0.7%#29 | 0.8%#28 | – | ARC Prize | |
| CritPt | 2.0%#181 | 9.4%#108 | 17.4%#70 | 22.9%#49 | 27.1%#31 | 30.0%#15 | – | Artificial Analysis | |
| Chess Puzzlesnot in index | 5.0%#156 | 22.0%#75 | – | – | – | 54.0%#9 | – | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | 14.0%#91 | 19.0%#59 | 10.0%#106 | – | – | 35.0%#19 | – | Epoch AI Benchmarking Hub | |
| LMCAnot in index | 46.5%#66 | 49.3%#54 | 49.6%#53 | 51.4%#45 | 52.2%#41 | 55.0%#32 | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | 78.7%#118 | 90.1%#66 | 90.1%#66 | 91.2%#57 | 92.5%#52 | 93.3%#49 | – | Epoch AI Benchmarking Hub |
Coding
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| SciCode | 44.6%#119 | 49.2%#92 | 49.6%#91 | 50.1%#87 | 51.6%#72 | 55.0%#41 | – | SciCode | |
| WeirdML | – | – | – | 78.3%#19 | – | – | – | WeirdML | |
| FrontierCode | – | – | – | – | – | – | 41.3%#20 | Cognition | |
| LMArena Coding | – | – | – | – | 1515#44 | – | – | LMArena | |
| LMArena WebDev | – | – | – | – | 1519#45 | – | – | LMArena | |
| LiveBench Codingnot in index | – | – | – | – | – | 78.3%#30 | – | LiveBench | |
| SciCode (AA)not in index | 45.1%#127 | 49.9%#102 | 50.5%#94 | 52.4%#77 | 52.3%#78 | 55.0%#50 | – | Artificial Analysis | |
| LiveCodeBench | – | – | – | – | 85.9%#25 | – | – | Vals AI | |
| IOI | – | – | – | – | – | 87.6%#7 | – | Vals AI | |
| SWE-bench (Vals)not in index | – | – | – | – | – | 95.4%#5 | – | Vals AI | |
| Code Migrationnot in index | – | – | – | – | – | 47.8%#11 | – | Vals AI | |
| ProgramBenchnot in index | – | – | – | – | – | 0.5%#13 | – | Vals AI | |
| Vibe Code Bench 1-100not in index | – | – | – | – | – | 14.8%#14 | – | Vals AI | |
| Vibe Code Bench v1.1not in index | – | – | – | – | – | 74.6%#29 | – | Vals AI | |
| Surface Evolver Benchnot in index | – | – | – | – | 83.8%#7 | – | – | Epoch AI Benchmarking Hub | |
| DeepSWEnot in index | – | 24.1%#66 | 35.1%#61 | 53.8%#42 | 60.2%#32 | 69.6%#14 | – | Epoch AI Benchmarking Hub | |
| CursorBenchnot in index | – | 25.2%#49 | 27.6%#48 | 30.7%#43 | 33.6%#34 | 41.3%#21 | – | Cursor | |
| ALE-Benchnot in index | – | – | – | – | – | 1951.4#7 | – | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in index | 0.5%#123 | 1.5%#106 | 1.0%#112 | 1.5%#106 | 10.1%#76 | 35.4%#31 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 56.2%#101 | 62.5%#86 | 72.3%#68 | 75.7%#60 | 80.2%#44 | 88.0%#12 | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| Terminal-Bench | – | – | – | – | – | 21.5%#63 | – | Terminal-Bench | |
| APEX-Agents | – | – | – | – | – | 58.2%#12 | – | Mercor | |
| LMArena Agent | – | – | – | – | 0.4#30 | – | – | LMArena | |
| LiveBench Agentic Codingnot in index | – | – | – | – | – | 55.0%#27 | – | LiveBench | |
| Terminal-Bench Hard | – | 43.9%#37 | – | 57.6%#11 | 62.9%#2 | 57.6%#11 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | – | 60.5%#178 | 72.8%#146 | 78.4%#129 | 80.4%#122 | 86.3%#88 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals) | – | – | – | – | – | 77.5%#14 | – | Vals AI | |
| GDPval (AA)not in index | 29.2%#127 | 29.8%#124 | 37.7%#95 | 43.1%#72 | 46.4%#58 | 47.7%#51 | – | Artificial Analysis | |
| τ²-Bench Banking (AA) | 15.7%#111 | 18.8%#99 | 25.6%#77 | 28.7%#69 | 29.7%#66 | 40.2%#29 | – | Artificial Analysis | |
| ITBench SRE (AA)not in index | – | – | – | – | – | 51.0%#5 | – | Artificial Analysis | |
| APEX-Agents (AA)not in index | – | – | – | – | – | 38.9%#4 | – | Artificial Analysis | |
| CyberBenchnot in index | – | – | – | – | – | 72.1%#14 | – | Vals AI | |
| Finance Agent v2not in index | – | – | – | – | – | 54.4%#21 | – | Vals AI | |
| Harvey's Legal Agent Benchmarknot in index | – | – | – | – | – | 0.8%#48 | – | Vals AI | |
| Legal Research Benchnot in index | – | – | – | – | – | 41.4%#20 | – | Vals AI | |
| SkillsBenchnot in index | – | – | – | – | – | 58.9%#11 | – | Vals AI | |
| Tax Agent Benchnot in index | – | – | – | – | – | 65.2%#17 | – | Vals AI | |
| Terminal-Bench 4.0 (Vals)not in index | – | – | – | – | – | 22.7%#18 | – | Vals AI | |
| Terminal-Bench Sciencenot in index | – | – | – | – | – | 10.0%#11 | – | Vals AI | |
| BALROGnot in index | – | – | – | – | – | 53.2%#6 | – | BALROG | |
| Vending-Bench 2not in index | – | – | – | – | – | – | 7343.2#11 | Andon Labs | |
| FACTS Searchnot in index | – | – | – | – | – | – | 81.7%#7 |
Maths
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | – | – | – | – | – | 86.0%#10 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | – | – | – | – | – | 70.7%#19 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 53.3%#170 | 88.9%#68 | – | – | – | 99.7%#12 | – | Epoch AI Benchmarking Hub | |
| ProofBench | – | – | – | – | 74.0%#12 | – | – | Vals AI | |
| LiveBench Mathematicsnot in index | – | – | – | – | – | 94.9%#17 | – | LiveBench | |
| LMArena Maths | – | – | – | – | 1479#35 | – | – | LMArena |
Knowledge
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| SimpleQA Verified | – | – | – | – | – | 43.2%#39 | – | Epoch AI Benchmarking Hub | |
| LiveBench Data Analysisnot in index | – | – | – | – | – | 79.3%#14 | – | LiveBench | |
| AA-Omniscience | -23.2#239 | -6.8#157 | -5.1#150 | -3.5#139 | -3.0#138 | 0.1#123 | – | Artificial Analysis | |
| MMLU-Pro | – | – | – | – | 86.7%#39 | – | – | Vals AI | |
| LegalBench | – | – | – | – | – | 85.1%#23 | – | Vals AI | |
| CorpFin | – | – | – | – | 65.3%#29 | – | – | Vals AI | |
| TaxEval | – | – | – | – | 76.2%#6 | – | – | Vals AI | |
| Excel Modeling Benchmarknot in index | – | – | – | – | – | 66.2%#20 | – | Vals AI | |
| MedCodenot in index | – | – | – | – | 43.4%#43 | – | – | Vals AI | |
| MedScribenot in index | – | – | – | – | 82.9%#46 | – | – | Vals AI | |
| Public Benefits Benchnot in index | – | – | – | – | – | 62.4%#23 | – | Vals AI | |
| GDP.pdfnot in index | – | – | 24.7%#12 | – | – | – | 24.7%#12 | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in index | 36.8%#139 | 43.7%#95 | 44.6%#91 | 45.5%#87 | 45.5%#86 | 46.8%#77 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 5.0%#500 | 10.1%#399 | 10.2%#393 | 10.2%#395 | 11.0%#380 | 12.1%#363 | – | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LiveBench Languagenot in index | – | – | – | – | – | 82.9%#21 | – | LiveBench | |
| LiveBench Instruction Followingnot in index | – | – | – | – | – | 64.6%#43 | – | LiveBench | |
| IFBench | – | 59.7%#125 | 62.2%#117 | 64.4%#106 | 66.3%#96 | 71.2%#58 | – | Artificial Analysis | |
| LMArena Instruction Followingnot in index | – | – | – | – | 1461#47 | – | – | LMArena |
Human preference
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LMArena Text | – | – | – | – | 1466#51 | – | – | LMArena | |
| EQ-Bench 4 | – | – | – | – | – | – | 1234#11 | EQ-Bench | |
| LMArena Creative Writingnot in index | – | – | – | – | 1426#76 | – | – | LMArena | |
| LMArena Multi-turnnot in index | – | – | – | – | 1470#61 | – | – | LMArena | |
| LMArena Longer Queriesnot in index | – | – | – | – | 1466#65 | – | – | LMArena |
Multimodal
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LMArena Vision | – | – | – | – | 1271#40 | – | – | LMArena | |
| MMMU-Pro | 66.7%#164 | 76.1%#88 | 76.8%#82 | 79.1%#60 | 79.5%#57 | 80.7%#42 | – | Artificial Analysis | |
| MMMU-Pro (Vals)not in index | – | – | – | – | – | 86.5%#20 | – | Vals AI | |
| MortgageTaxnot in index | – | – | – | – | 67.3%#26 | – | – | Vals AI | |
| SAGEnot in index | – | – | – | – | – | 47.0%#33 | – | Vals AI | |
| Blueprint-Bench 2not in index | – | – | – | – | – | – | 30.8%#12 | Epoch AI Benchmarking Hub |
Long context
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| AA-LCR | 58.7%#268 | 71.3%#188 | 74.0%#153 | 77.7%#116 | 79.0%#97 | 83.0%#24 | – | Artificial Analysis | |
| LMArena Documentnot in index | – | – | – | – | 1477#14 | – | – | LMArena |
Composite
| Benchmark | no reasoning | low | medium | high | xhigh | max | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in index | – | – | – | – | – | – | 159.8#9 | Epoch AI Benchmarking Hub | |
| LiveBench | – | – | – | – | – | 77.9%#20 | – | LiveBench | |
| AA Intelligence Index | 20.8#205 | 27.5#128 | 30.1#108 | 34.2#84 | 38.0#70 | 42.1#48 | – | Artificial Analysis | |
| Vals Indexnot in index | – | – | – | – | – | 53.1#19 | – | Vals AI | |
| Vals Multimodal Indexnot in index | – | – | – | – | 65.1%#14 | – | – | Vals AI |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Azure | 84 tok/s | 3.66 s | $2.00 | $12.00 | 1.1M | – |
| OpenAI | 79 tok/s | 2.51 s | $2.00 | $12.00 | 1.1M | – |
| OpenAI | 72 tok/s | 1.73 s | $1.00 | $6.00 | 1.1M | – |
| Amazon Bedrock | 58 tok/s | 1.73 s | $2.20 | $13.20 | 1.1M | – |
| OpenAI | 56 tok/s | 1.01 s | $4.00 | $24.00 | 1.1M | – |
Price history
Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.200 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0044 | $0.0039 | 3.0 min |
| Summarise a 30-page report | 12,000 / 600 | $0.031 | $0.015 | 3.0 min |
| Code edit | 6,000 / 1,500 | $0.030 | $0.022 | 3.2 min |
| Agentic coding session | 60,000 / 4,000 | $0.168 | $0.087 | 3.6 min |
| Structured extraction | 2,000 / 200 | $0.0064 | $0.0037 | 3.0 min |
See also
Data as of 4 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “GPT-5.6 Terra: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/gpt-5-6-terra, data as of 4 Oct 2026.