`anthropic-sdk-python` `v0.105.0` adds `claude-opus-4-8` support

// related

• #0001

  #0001Models & APIGeekNews22 hours ago

  Anthropic Ships `Claude Opus 4.8`

  90radar

  Same-price upgrade over 4.7, with better benchmarks and collaboration. Worth re-testing as a coding and agent default before changing spend assumptions.

  • —Built on Claude Opus 4.7; this is an incremental top-tier model refresh, not a new product line.
  • —Benchmark performance improved while pricing stays unchanged. Existing API cost models do not need recalculation first.
  • —Collaboration capability was strengthened, so agent workflows and pair-programming style tasks deserve fresh evals.

  Source: news.hada.io/topic?id=29960Read original →
• #0002

  #0002Models & APIAnthropic TypeScript SDK24 hours ago

  `Anthropic TypeScript SDK` `0.100.0` Adds `claude-opus-4-8` Support

  70radar

  Anthropic TypeScript SDKTypeScript SDK — typed client for Anthropic APIs

  Model access, mid-conversation system blocks, and detailed output-token usage landed together. Worth upgrading before testing new Claude flows.

  • —claude-opus-4-8 is now supported in the SDK, so TypeScript apps can call it without local type workarounds.
  • —Mid-conversation system blocks unlock cleaner control changes during long chats and agent loops.
  • —usage.output_tokens_details adds more granular output accounting, useful for cost dashboards and per-feature metering.

  Source: github.com/anthropics/anthropic-sdk-typescript/releases/Read original →
• #0003

  #0003Models & APIvercel_blogyesterday

  `Claude Opus 4.8` lands on `Vercel AI Gateway`

  80radar

  Vercel AI GatewayLLM gateway — unified calls, cost tracking, and failover

  Long-running coding and knowledge tasks can now route through one Vercel API. Useful if you already want retries, cost tracking, and failover without wiring providers yourself.

  • —Use anthropic/claude-opus-4.8 in the AI SDK; model switching stays inside the existing Gateway path.
  • —Positioned for long-horizon agent work: multi-step refactors and complex coding tasks need less mid-task correction.
  • —Gateway adds usage and cost tracking, retries, failover, and latency/cost-based provider sorting. Less infra glue around LLM calls.
  • —Pricing mirrors provider rates with no Vercel markup or platform inference fee, including BYOK requests.

  Source: vercel.com/changelog/opus-4-8-on-ai-gatewayRead original →