Models & API
GPT-5.6 family selection and the OpenAI API in practice: Responses API, streaming, function calling, billing.
2 channels · 25 articles
GPT Models
8 articlesThe GPT-6 family grows: gpt-6-sol, gpt-6-luna, gpt-6.1-sol land in the official SDKs
Verified 2026-10-09: openai-python and openai-node enums both list gpt-6-astra, gpt-6.1-sol, gpt-6-sol, gpt-6-luna; official RSS positions Sol/Luna and GPT-6.1 Sol. Only SDK- and same-day-RSS-verifiable facts are asserted.
Read articleGPT-6-Astra Goes Live: Codex Model Picker and Amazon Bedrock Catalogs
Codex CLI 0.154.0 (2026-09-09) lists GPT-6-Astra first in its model picker and adds it to Amazon Bedrock; a bundled official guide confirms Responses API usage and six reasoning levels topped by ultra. Pricing figures remain unpublished.
Read articlegpt-6-astra appears in the OpenAI SDK: the new ChatModel ID and the Safety Alerts API
openai-python v3.8.0 and openai-node v7.10.0 (2026-09-03) added gpt-6-astra to the ChatModel enum ahead of the GPT-5.6 tiers, plus a Safety Alerts API and misalignment error details. This article only states what SDK source shows.
Read articlereasoning.effort in Practice: Making Reasoning Depth a Tunable Parameter
Same model, adjustable thinking depth. This tutorial covers the reasoning.effort syntax, per-tier scenarios, how it interacts with max_output_tokens and the incomplete status, and the cost consequences of getting it wrong.
Read articleGPT-5.6 vs Claude 4.5 Sonnet vs Gemini 2.5 Pro: 2026-08 production comparison
GPT-5.6 / Claude 4.5 Sonnet / Gemini 2.5 Pro flagship models compared: code / multimodal / long context / long reasoning / tool use / price. Third-party benchmarks + real scenario tests. Multi-model selection decision matrix.
Read articleGPT-5.6 Fine-Tuning in Practice: Cost, Quality, and When to Skip It
GPT-5.6 fine-tuning (SFT / DPO): when it's worth doing and when prompt engineering is cheaper. A decision framework, cost estimate, and three real scenarios.
Read articleGPT-5.6 Model Selection Guide (2026): Sol, Terra, or Luna?
How to pick between GPT-5.6 Sol / Terra / Luna without wasting money or breaking quality: a workload-first framework, the reasoning.effort dial, cost math, and the o-series boundary.
Read articleThe complete guide to GPT models (2026-07): GPT-5.6 Sol, Terra, Luna
OpenAI's current flagship model family is GPT-5.6 (2026-07-09), split into Sol / Terra / Luna. Capability tiers, pricing, context, reasoning effort, and which to pick — plus GPT-Realtime-2.1 voice and GPT Image 2.
Read articleAPI
17 articlesopenai-python 3.32.0 and openai-node 7.35.0: prompt-created voices, Live audio fixtures, and Decisions in Live
Both official SDKs shipped on 2026-10-16: voices created from a text description (Live only), 30-second Live audio fixtures, Decisions transcript triggers, and prompt caching for the Decisions API - all from the CHANGELOGs and tag sources.
Read articleBrowser and Computer Toolsets (1/4) Land in the OpenAI SDKs: BrowserToolset Turns Driver Methods into Function Tools
On 2026-10-14 the official SDKs shipped four releases: openai-python 3.30/3.31 and openai-node 7.33/7.34. The (1/4) browser and computer toolsets headline: BrowserToolset wraps a caller-supplied driver as ordinary function tools.
Read articleHosted agent environments surface in the OpenAI SDKs: prewarmed creation, idle suspension, expiration
Four releases in three days, October 9-11, 2026: openai-python 3.27-3.29 and openai-node 7.32 surface one agent environment API — prewarmed beta creation, idle suspension, expiration. Quoted from release notes and SDK source.
Read articleOpenAI API 429 Rate Limit Errors: RateLimitError and SDK Retries Explained
A 429 is two problems in one status: throughput limits vs quota exhaustion. The Python SDK already retries twice and honors Retry-After — this guide explains the mechanics from source.
Read articleopenai-node v7.20.0 Explained: Environment-Variable Vault Credentials, External Storage, and Safety Cases
Six PRs in one openai-node release: environment_variable vault credentials, external storage management, safety case retrieval with two webhook events, a SIP media security field, and a legacy GET fix — each traced to PR and tag sources.
Read articleopenai-python 3.15/3.16 and openai-node 7.18/7.19: Cache Prewarming, Webhook Management, connector_id Deprecation
Six OpenAI SDK releases in one day: prewarm cache warming, client.webhooks endpoint management, connector_id deprecated for post-September-1 models, WebSocket sessions in both languages. Every item traced to its PR.
Read articleThe Responses API Compaction Progress Event: response.compaction.compacting and compaction_trigger
openai-node v7.17.0 adds a Responses API compaction progress event: response.compaction.compacting fires at most once every 30 seconds and carries no summary content. Set in context: trigger item, compact endpoint, context_management.
Read articleopenai-python 3.14.x and openai-node 7.16/7.17: Stream Error Normalization, WebSocket Backpressure, and SSE Fixes
Four OpenAI SDK releases in three days, one theme: reliability — Python normalizes stream errors into SDK exceptions, stringifies error codes, checks max_retries; Node caps WebSocket iterator backpressure and fixes lost terminal SSE events.
Read articleThe Agents API appears in the OpenAI SDK (beta): /agents CRUD, environments, sessions, and vaults
openai-python v3.13.0 / openai-node v7.15.0 (09-10) add the Agents API (beta): reusable Agent CRUD, managed sessions with execution environments, subagents, and credential vaults. Only SDK-source-verifiable facts.
Read articleOpenAI API Key Expiration: expires_in_seconds, Org Policies, and Safe Automation
Service Account API key expiry got its key upgrade in v3.11.0 / v7.13.0: org and project policies can now force expiration with a hard maximum lifetime. Parameter boundaries, code, and an ops checklist, all in one place.
Read articleopenai-python 3.9/3.10 and openai-node 7.11/7.12: Prompt Cache Diagnostics, API Key Expiry, GPT Image 2.5
Four SDK releases in two days, two more for key governance, and two pairs on 09-10: prompt cache diagnostics, key expiry plus org policy, GPT Image 2.5 — and two brand-new API surfaces (Live, Agents). Itemized, with code.
Read articleResponses API vs Chat Completions: Is It Time to Migrate?
The Assistants API is shut down and Chat Completions is in legacy mode. This guide compares the two generations across state management, tool definitions, and built-in capabilities, with a migrate-one-flow-at-a-time path.
Read articleOpenAI Structured Outputs: The Complete Guide to json_schema, strict Mode, and Common Errors
The full path to reliably parseable JSON: the Responses API text.format syntax, strict mode, schema design tips, and troubleshooting for high-frequency errors like passing schema without name.
Read articleResponses API advanced: structured outputs, streaming SSE, Batch API, prompt caching
Responses API advanced usage: JSON Schema strict mode, streaming SSE parsing, Batch API offline cost-cut, prompt caching three-tier cache, cost optimization case studies. From 'it calls' to 'production-grade'.
Read articleOpenAI API Error Handling and Retry: 401/429/5xx Patterns
Production-ready error handling for the OpenAI API: 401/429/500/503/timeout. Exponential backoff + jitter, error budgets, upstream protection, streaming-mode specifics.
Read articleFunction Calling with the OpenAI API: A Complete Guide to Tool Use in the Responses API
Function calling is how GPT-5.6 calls your code. This guide walks the full Responses API flow: declaring tools, parsing function_call, returning results, and chaining multi-turn tool calls.
Read articleOpenAI API Beginner: Your First GPT-5.6 Call Explained
From account creation and API Key retrieval to your first Responses API call in Python and Node.js — a complete beginner's path to the OpenAI API.
Read article