What is Gpt 4 Turbo Preview?
Gpt 4 Turbo Preview is a language model in the o10 fulfillment catalog (slug: openai-gpt-4-turbo-preview). Published input pricing starts at $10.00/1M tokens. Context window: 128K tokens.
Catalog snapshot 2026-06-27.
Gpt 4 Turbo Preview costs $10.00/1M input tokens and $30.00/1M output tokens as of 2026-06-27. Openai publishes this model with a 128K tokens context window. o10 lists Gpt 4 Turbo Preview on OpenRouter at routing band C3.
Last updated: 2026-06-27. Full catalog
Gpt 4 Turbo Preview — specs, hosts, and routing
AnswersSelf-contained
How much does Gpt 4 Turbo Preview cost per million tokens?
Gpt 4 Turbo Preview costs $10.00/1M input tokens and $30.00/1M output tokens in the gateway catalog snapshot (2026-06-27). Endpoint pricing may differ by fulfillment host.
Sourced from gateway catalog snapshot.
Where can you run Gpt 4 Turbo Preview?
Gpt 4 Turbo Preview is configured on OpenRouter. Each host maps to an upstream model ID listed in the fulfillment routes table. This model is routable through o10 eval-gated routing.
What is Gpt 4 Turbo Preview's context window?
Gpt 4 Turbo Preview supports a 128K tokens context window in the gateway catalog snapshot. Context limits may vary by fulfillment host endpoint.
What o10 routing band is Gpt 4 Turbo Preview?
Gpt 4 Turbo Preview is classified at o10 routing band C3. Bands group models by cost tier and eval profile: C0 (economy), C1 (standard), C2 (capable), C3/C4 (frontier). Production routing is enabled for this slug.
Who makes Gpt 4 Turbo Preview?
Gpt 4 Turbo Preview is attributed to source provider openai in the o10 catalog. Fulfillment hosts (OpenRouter) supply API access; list pricing may differ by venue.
01Deep dive
Specifications
Gpt 4 Turbo Preview in the o10 unified catalog.
| Field | Value |
|---|---|
| o10 slug | `openai-gpt-4-turbo-preview` |
| Type | text |
| Band | C3 |
| Routable | Yes |
| Source provider | openai |
| Primary host | OpenRouter |
| Gateway ID | `openai/gpt-4-turbo` |
| Context window | 128,000 tokens |
| Released | 2023-11-06 |
02Deep dive
Token pricing
Gateway snapshot pricing for Gpt 4 Turbo Preview.
| Unit | Price |
|---|---|
| Input | $10/1M tokens |
| Output | $30/1M tokens |
03Deep dive
Fulfillment hosts
Upstream venues configured for Gpt 4 Turbo Preview.
| Host | Upstream model ID |
|---|---|
| openrouter | openrouter/openai/gpt-4-turbo-preview |
04Deep dive
Gateway provider endpoints
Endpoint-level pricing and latency from the gateway catalog.
| Provider | Input ($/1M) | Latency p50 | Uptime 1d |
|---|---|---|---|
| openai | $10 | 1027.5ms | 10000.0% |
05Deep dive
Routing with o10
Eval-gated model selection for this slug.
o10 slug: openai-gpt-4-turbo-preview. Band: C3. Routable: yes.
See [/models/routing](/models/routing) for smart aliases (o10/auto, o10/economy) and [/kyi](/kyi) for governance.
SourceMethodology
Fulfillment catalog 2026-06-27. Gateway pricing joined where slug match exists.
FAQFrequently asked questions
Common questions
How much does Gpt 4 Turbo Preview cost?
Gpt 4 Turbo Preview costs $10.00/1M input tokens and $30.00/1M output tokens in the gateway snapshot (2026-06-27). Verify against your provider's published pricing.
What is Gpt 4 Turbo Preview's context window?
Gpt 4 Turbo Preview supports 128K tokens in the gateway catalog snapshot. Host-specific endpoints may differ.
Which hosts serve Gpt 4 Turbo Preview?
Gpt 4 Turbo Preview (openai-gpt-4-turbo-preview) is configured on: OpenRouter. Upstream model IDs are listed in the fulfillment routes section.
Is Gpt 4 Turbo Preview routable through o10?
Yes. Gpt 4 Turbo Preview is routable at band C3. Send model: "openai-gpt-4-turbo-preview" or use an o10 smart routing alias.
How do I call Gpt 4 Turbo Preview via o10?
Use o10 slug "openai-gpt-4-turbo-preview" or gateway ID "openai/gpt-4-turbo" depending on your integration. Gateway-compatible endpoints accept OpenAI-style /v1 requests.
How does o10 route this model?
o10 routes Gpt 4 Turbo Preview only when per-use-case evals clear at your quality floor. Band C3 indicates cost tier. See /kyi for governance and /models/routing for smart aliases.