OpenAI
Chat with GPT-5.5 online
OpenAI's frontier model for the long, careful jobs — a million tokens of context, tools, vision, and time to think.
1.05M tokens
Context
128,000 tokens
Max output
Deliberate
Speed
Straight from GPT-5.5
Three prompts, three unedited answers, generated on September 2, 2026 through NinjaChat. Click the prompt to copy it and run it yourself.
Write a Python function that merges overlapping intervals, plus three short tests. Keep the whole answer under 40 lines of code.
def merge_intervals(intervals):
if not intervals:
return []
intervals = sorted(intervals)
merged = [list(intervals[0])]
for start, end in intervals[1:]:
if start <= merged[-1][1]:
merged[-1][1] = max(merged[-1][1], end)
else:
merged.append([start, end])
return [tuple(interval) for interval in merged]
assert merge_intervals([(1, 3), (2, 6), (8, 10)]) == [(1, 6), (8, 10)]
assert merge_intervals([(5, 7), (1, 2), (2, 4)]) == [(1, 4), (5, 7)]
assert merge_intervals([]) == []
Read: We ran the same 3 prompts on 12 AI models — Unedited answers, latency and price for every model, measured through the NinjaChat API in September 2026.
Served by OpenAI as gpt-5.5. Answers vary run to run; these are one take each, not the best of several.
What you’ll use it for
Complex analysis
Plans, reviews and decisions that need many careful steps.
Large codebases
Multi-file changes with edge cases, run with tools.
Long documents
More than a million tokens of context in one thread.
Vision and files
Screenshots, diagrams, PDFs and spreadsheets in the conversation.
Prompts to steal
Tuned to this model — click any line to copy.
About GPT-5.5
GPT-5.5 is the frontier model OpenAI released on April 24, 2026, built for complex professional workloads: multi-step analysis, large codebases, long documents and agent-style tasks where the model has to plan, use tools and check its own work. On NinjaChat it is included in every plan, alongside GPT-5.6 and the rest of the catalog, with no ChatGPT subscription needed.
What makes 5.5 different from the GPT-5.4 family is headroom. The context window runs past a million tokens with 128K tokens of output, so a whole repository, a contract set or a quarter of research fits in one conversation. It reads images and files, calls tools, and returns structured JSON when you ask, which is why it works as well inside a workflow as it does in a chat window.
In practice GPT-5.5 is the model for the task you would hand a senior colleague: a refactor with edge cases, a plan that has to survive scrutiny, a document that needs arguing with rather than summarizing. It is deliberate rather than quick. For everyday questions a faster model feels better, and NinjaChat keeps both a click apart.
Since July 2026 OpenAI's newest generation is GPT-5.6, split into Sol, Terra and Luna tiers. GPT-5.5 sits between: more capable than Luna and Terra on the hardest problems, priced like Sol on the API, and still the right pick when your prompts and evaluations were tuned for it. Both generations are on NinjaChat, so the practical test is to run the same prompt on each.
How to use GPT-5.5
From zero to your first result in under a minute.
01
Create a NinjaChat account and choose a plan
02
Open chat and pick GPT-5.5 from the model list
03
Give it the whole task and the files, not a fragment
04
Ask for a plan first on big jobs, then say go
05
Compare the answer with GPT-5.6 Sol or Claude Opus 5 in one click
Tips for better results
- Use 5.5 for the hard part and a faster model for follow-ups
- Be explicit about output format; it honors it
- Attach the real files instead of describing them
- For anything recent, turn on search
GPT-5.5 vs the alternatives
Where it wins, and where it doesn't.
vs GPT-5.6 Sol
Prompts tuned for 5.5 behave exactly as before
Sol is the newer deep-reasoning tier
vs GPT-5.6 Luna
ViewFar more depth on hard, multi-step work
Luna is much faster and cheaper for everyday tasks
vs Claude Opus 5
ViewOpenAI's tool ecosystem and a larger context window
Opus 5 leads on agentic coding benchmarks
Questions, answered.
Beyond chat — images & video
The same subscription covers the image studio and video generation.
GPT-5.5, and 50 more.
One subscription covers every model on NinjaChat, GPT-5.5 included.
