← Vaquero

Q3 2023 — Q2 2026  /  36 models  +  Q3 2026 to date

How much can it hold?
Twelve quarters of frontier context windows.

One bar per quarter for each of three camps: OpenAI's flagship, Anthropic's flagship, and the leading Asian lab's flagship. Values are the maximum advertised context window of the most capable model each camp had shipped by the end of that quarter. Hover any bar for the details. The twelve solid columns are completed quarters through 30 June 2026; the faded thirteenth is Q3 2026 so far, current to 25 July.

OpenAI
Anthropic
Leading Asian lab
Scale
Hover a bar to read the model, the token count, and the caveat.

How models were picked

Flagship, not largest window. Each bar is the camp's most capable model that quarter, then its context window — not the biggest window the camp could field. That's why the line isn't always upward.

Maximum advertised. Where a window is tiered, the top tier is plotted and flagged in the readout. GPT‑5.4 is 272K standard and 1M through the API and Codex; the 1M figure is plotted.

"Leading Asian lab" is a judgment call. No single Chinese lab held the frontier throughout, so the pick moves between Alibaba, Moonshot and DeepSeek. Reasonable people would draw some quarters differently.

Advertised ≠ effective. Every model benchmarked so far degrades on recall well before its stated limit.

What the chart shows

Two step changes, not a smooth curve. 32K to 128K over late 2023, then a long plateau at 128K–200K that lasts nearly two years, then everyone lands on 1M inside a single quarter.

Anthropic sat at 200K for nine straight quarters — Claude 2.1 through Opus 4.5 — before Opus 4.6 moved the Opus line to 1M in February 2026.

Q1 2026 is the convergence point. GPT‑5.4, Opus 4.7 and Qwen 3.6 Plus all advertise roughly 1M within weeks of each other, and the number stops being a differentiator.

Q2 2026 dips on the Asian bar because the quarter's strongest Asian model, Kimi K2.6, ships 256K. MiniMax M3 hit 1M the same quarter but wasn't the capability leader.

Three weeks into Q3 2026, the gap has closed entirely. Kimi K3 (16 July), Claude Opus 5 (24 July) and GPT‑5.6 Sol all sit at 1M or just over. Every camp is now flat at the same number, and the next axis of competition is effective recall and price per filled window, not the headline figure.

Sources

  1. Anthropic, Context windows — Claude Platform Docs. Confirms 1M for Opus 4.6/4.7/4.8/5, Sonnet 4.6/5, Fable 5, Mythos 5; 200K for Sonnet 4.5 and earlier.
  2. Anthropic, Introducing Claude Opus 5, 24 July 2026 — state of the art on Frontier-Bench and GDPval‑AA, behind Mythos 5 on cybersecurity. Developer platform notes give 1M context and 128K output at Opus 4.8 pricing.
  3. Anthropic, Introducing Claude Sonnet 4.6.
  4. innFactory, Moonshot Kimi model tracker — Kimi K3, 16 July 2026, 2.8T parameters, 1M context, debuted third on the Artificial Analysis leaderboard.
  5. OpenAI, Models — API docs. GPT‑5.6 Luna / Terra / Sol at 1.05M context, 128K max output, Feb 16 2026 cutoff.
  6. Google Cloud Vertex AI, Claude Opus 4.6 and Claude Opus 4.5 model pages — 1,000,000 vs 200,000 max input tokens.
  7. Morph, LLM Context Window Comparison (2026), June 9 2026 — 1M tier for Fable 5, Opus 4.8, GPT‑5.5, GPT‑5.4.
  8. Digital Applied, AI Context Window Comparison 2026 — GPT‑5.4 at 272K standard / 1M via API and Codex; Qwen 3.6 Plus at 1M.
  9. Digital Applied, GPT‑5.2 and Codex — 400K context, 128K output, released Dec 11 2025.
  10. Wikipedia, GPT‑5.2, GPT‑5.1, GPT‑5.4, GPT‑4.5 — release dates and lineage.
  11. Wikipedia, Kimi (chatbot) — first public Kimi at 128K lossless context, Nov 16 2023; 2M-character beta, March 2024.
  12. DeepInfra, Kimi K2.6 API benchmarks and Verdent, What is Kimi K2.6 — 262,144-token window, released April 20 2026.
  13. Atlas Cloud, Kimi K2.6 vs GLM 5.1 vs Qwen 3.6 Plus vs MiniMax M2.7 — Qwen 3.6 Plus as the only 1M-context model in that group, late March 2026.
  14. GEO Toolbox, Chinese AI Models Compared (2026) — MiniMax M3 1M sparse-attention window, June 1 2026; DeepSeek V4 timeline.
  15. Index.dev, Top Chinese AI Models — Qwen3‑Max at 262K, Kimi K2 Thinking, GLM‑4.6 at 200K.
  16. hidekazu-konishi.com, Anthropic Claude Model Release Timeline — Claude 100K first (May 2023), Claude 2.1 200K, generation dates.
  17. NVIDIA et al., ChatQA 2 (arXiv:2407.14482) — Yi‑34B extended to 200K, Qwen2‑72B extrapolated to 128K.
  18. Simon Willison, The new GPT‑5.6 family and GPT‑5.2.

Compiled 25 July 2026. Pre‑2025 figures are from vendor model cards and release posts at the time.