Data as of October 25, 2026 · How the score is built
GPT Realtime 2.1
Decision readingGPT Realtime 2.1 is tracked, but not publicly ranked yet. Published speech or voice results from Audio Realism Benchmark appear below. They do not produce a weighted text-model score.
Released Jul 6, 2026 — see all recent releases
Voice benchmark evidence
Audio Realism Benchmark is an owner-defined voice evaluation. Its result stays separate from BenchLM's weighted text-model ranking.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- gpt-realtime-2.1OpenAI GPT Realtime 2.1 model documentation
- Context window
- 128KOpenAI GPT Realtime 2.1 model documentation
- Maximum output
- Not sourced yet
- Knowledge cutoff
- September 30, 2024OpenAI GPT Realtime 2.1 model documentation
- Input modalities
- text, audio, imageOpenAI GPT Realtime 2.1 model documentation
- Output modalities
- text, audioOpenAI GPT Realtime 2.1 model documentation
- Parameters
- Not disclosed by the provider
- Availability
- OpenAI APIOpenAI GPT Realtime 2.1 model documentation
- Cloud regions
- Not tracked yet
- Lifecycle
- activeOpenAI GPT Realtime 2.1 model documentation
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.40 per million cached input tokensOpenAI model documentation
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track GPT Realtime 2.1, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
GPT Realtime 2.1 is a proprietary model with a 128K context window. No explicit reasoning mode is documented in this profile.
Available on the OpenAI Realtime API (v1/realtime over WebRTC, WebSocket, or SIP) as gpt-realtime-2.1 since July 6, 2026, with text, audio, and image input, text and audio output, a 128,000-token context, and 32,000-token maximum output. OpenAI reports a 25% p95 latency reduction across Realtime models with this release.
GPT Realtime 2.1 is a display-only speech-to-speech reasoning model with configurable reasoning effort and tool use on the Realtime API. Voice-protocol results remain separate from BenchLM’s weighted text-model ranking.
Speech or voice evidence appears above; text-model benchmark coverage remains empty.
Last updated October 25, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does GPT Realtime 2.1 perform overall in AI benchmarks?
GPT Realtime 2.1 has published speech or voice results in Audio Realism Benchmark. Those results remain separate from the weighted text-model ranking, so this profile does not assign a BenchLM score or overall rank. Use the linked evaluation's metric and run conditions to compare speech systems.
Does GPT Realtime 2.1 have full benchmark coverage on BenchLM?
No. GPT Realtime 2.1's speech or voice results appear in Audio Realism Benchmark, but it has no published text-model benchmark rows. The voice evaluation does not fill the text-model coverage slots or produce an overall rank. Missing text coverage is not treated as a zero score.
What is the context window size of GPT Realtime 2.1?
GPT Realtime 2.1 has a documented context window of 128K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.