OpenAI Launches GPT-6.1 Terra: Computer-Use Agent Model with 50% API Price Cut

On October 27, OpenAI officially released GPT-6.1 Terra, the first Terra-class model since GPT-5.6, positioned as "faster, stronger, and cheaper computer use": it can understand screen content, interpret screenshots, perform actions across apps, and write naturally, targeting full-workday, high-throughput agent scenarios. The model is now fully available to Plus, Pro, Enterprise, and Edu users in ChatGPT Work and Codex; developers can call it via API as gpt-6.1-terra, and it is already live on platforms like LMArena and Design Arena.

Confirmed

• Pricing: $1/$5 (per million input/output tokens), half or less of GPT-5.6 Terra ($2/$12) — a 50% cut on standard API prices, with output prices down roughly 58%.
• Official benchmarks: about 15 points above GPT-5.6 Terra on Agents' Last Exam v1 (which tests agents using complex professional software), scoring 56%-59% at under $1 in cost, matching GPT-6 Astra at 1/10 the per-token price while completing tasks over 3x faster.
• OpenAI also claims that at max effort, token efficiency is 3x that of Claude Opus 5.5, with per-task cost about 1/9.
• Per comparison data relayed by markk, GPT-6.1 Terra ranked first across six benchmarks (including SWE-bench 76.2%) against GPT-5.6 Terra, GPT-6 Astra, GPT-6.1 Sol, Claude Opus 5.5, and Gemini 2.5 Pro; however, m7's benchmark chart shows Claude Opus 5.5 still leads on Agents' Last Exam.

Unconfirmed

• bindureddy called it "Sonnet-class at about half the price of Claude Sonnet," and scaling01's price-cut claims came from a third-party account — both are personal accounts.

Why it matters

• Terra continues OpenAI's cost-efficiency-focused strategy, bringing GPT-6-level agent capability to the $1 per-task range, potentially lowering the barrier to building computer-use agents significantly.
• Community reactions are split: markk found the release disappointing, noting the Terminal-Bench 4.0 chart shows it trailing Claude Sonnet-class models at high cost, and quipped that "AI companies taken over by doomers means a flash-frozen AI winter" — revealing a gap between the official narrative and some users' real-world expectations.

2026-10-27 ~ 2026-10-27 · 34 related posts

Full story(4 episodes)→

• Episode 1: Microsoft PR leak reveals unreleased OpenAI model GPT-6.1 Terra(2026-10-19, 6 posts)
• Episode 2: GPT-6.1 Terra Surfaces Across Microsoft, GitHub, and OpenAI API Channels(2026-10-25, 9 posts)
• Episode 3: Report: OpenAI pauses frontier RL training, shifts to smaller cheaper models(2026-10-25, 2 posts)
• Episode 4: OpenAI Launches GPT-6.1 Terra: Computer-Use Agent Model with 50% API Price Cut(2026-10-27, 34 posts)

• GPT-6.1-Terra Live With Big Price Cut: $5 vs $12 Output Pricing — scaling01 · 2026-10-27
• GPT-6.1 Terra API pricing revealed: $1 input / $5 output, $0.05 cached — airesearch12 · 2026-10-27
• OpenAI launches GPT-6.1 Terra with 50% lower API prices than GPT-5.6 Terra — OpenAI · 2026-10-27
• [source] GPT-6.1 Terra matches GPT-6 Astra at 1/10th the price, 3x faster on agent tasks — OpenAI · 2026-10-27
• [source] GPT-6.1 Terra rolls out to all Plus, Pro, Enterprise and Edu users — OpenAI · 2026-10-27
• OpenAI launches GPT-6.1 Terra: faster, cheaper computer-use model for the whole workday — airesearch12 · 2026-10-27
• Leaked Terminal-Bench 4.0 chart shows GPT-6.1 Terra leading at lowest cost, unverified — Angaisb_ · 2026-10-27
• GPT-6.1 Terra arrives, trained for fast and efficient computer use — mark_k · 2026-10-27
• OpenAI launches GPT-6.1 Terra, a cost-efficient agent-focused model variant — daddyhughes111 · 2026-10-27
• OpenAI ships GPT-6.1 Terra: +15 points on Agents' Last Exam at roughly 1/3 the cost — reach_vb · 2026-10-27
• GPT-6.1 Terra lands: early users praise speed, value and subagent performance — BenBajarin · 2026-10-27
• ChatGPT 6 Terra benchmark charts surface, showing score-cost curves — 141_1337 · 2026-10-27
• GPT-6.1 Terra benchmarks: strong score under $1, but Claude Opus 5.5 still leads — 141_1337 · 2026-10-27
• GPT-6.1 Terra Benchmarks: Sweeps All Six Tests, SWE-bench at 76.2% — mark_k · 2026-10-27
• OpenAI Launches GPT-6.1 Terra at $1/M Input Tokens, ~9x Cheaper per Task Than Opus 5.5 — VraserX · 2026-10-27
• GPT-6.1 Terra Now Testable on LMArena in Battle and Agent Modes — arena · 2026-10-27
• OpenAI lists GPT-6.1 Terra: 1.05M context, $1/$5 pricing — but this poster sticks with Opus 5.5 — BLCNYY · 2026-10-27
• GPT-6.1 Terra model card: 1.05M context, $1/$5 pricing, six effort tiers — TinfoilTricorn · 2026-10-27
• GPT-6.1 Terra reportedly launches at $1/M input, half the price of Sonnet — bindureddy · 2026-10-27
• Unverified: GPT-6.1 Terra reportedly released, leaked chart shows top score per dollar on Agents' Last Exam — ZeroStateReflex · 2026-10-27
• GPT-6.1 Terra: Near-Astra Performance at ~9x Cheaper per Task Than Opus 5.5 — shaunralston · 2026-10-27
• Chart: GPT-6.1 Terra beats Sol max on cost, and max variant outperforms Astra max — GabGarrett · 2026-10-27
• OpenAI ships GPT-6.1 Terra, but Claude Sonnet 5.5 beats it on Terminal-Bench 4.0 — imjustnewatai · 2026-10-27
• OpenAI launches budget computer-use model GPT-6.1 Terra, but why choose it over Sol? — lyraxana · 2026-10-27
• 'Doomer takeover = AI winter': Terra 6.1 flops on Terminal-Bench 4.0 vs Sonnet 5.5 — mark_k · 2026-10-27

9 near-duplicate retellings: OpenAIDevs · OpenAIDevs · ArtificialAnlys · Angaisb_ · OpenAI · testingcatalog · testingcatalog · ChrisGPT · soleio