Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-09-02 · high
Cline switched 57 providers to it within a day, and forcing a specific tool call now returns a 400 error.
Why now: Anthropic released Fable 5.1 on September 1, and by September 2 the default swap had already reached Bedrock, Vertex, OpenRouter and DigitalOcean.
Claude Fable 5.1 became the default coding model across most major agent tools within a day of its September 1 release. Most harnesses don't pin a model version. A script that called Claude on August 31 may hit different rules and costs by September 2.
Anthropic's Claude Code set Fable 5.1 as the default Fable model first, switching in v2.1.257 on September 1, the claude-code CHANGELOG shows. The model runs a 1M-token context window at $10 per million input tokens and $50 per million output.
By September 2, Cline v4.1.17 had moved its default across all 57 providers, according to Cline's release notes. Bedrock, Vertex, OpenRouter and DigitalOcean all switched from Opus 5 to Fable 5.1 together.
Copilot CLI v1.0.83-2 added support for claude-fable-5.1. GitHub retired six Copilot models, including Opus 4.5 and 4.6.
Two changes break existing code with no version bump. Forcing a specific tool call, tool_choice of type any or a named tool, now returns a 400 error on Fable 5.1. Thinking runs on every turn, and a forced call skips it, pushing reasoning into the tool arguments instead.
The fix is setting tool_choice to auto with strict enabled, per Anthropic's Fable 5.1 docs.
Thinking blocks are also bound to the conversation prefix. Editing a system prompt, tool list, or earlier message returns a 400 for belonging to a different conversation. Enforcement applies to accounts created on or after August 31, 2026.
opencode 1.18.26 shipped a compatibility patch so sessions tolerate stale thinking blocks instead of failing the turn.
These regressions don't need a code change to show up. Anthropic warns Fable 5.1 may issue one tool call per turn where Fable 5 batched several. It's more likely to rewrite a whole file than make a targeted edit.
Simon Willison's one-shot SVG test at Max thinking cost $3.30 and took 14 minutes. Latent Space measured a 75% cut in cache-read pricing against a 20% higher per-task cost, since it emits 1.7 times the output tokens.
Every Fable 5.1 output also carries Anthropic's statistical text watermark, the first time a Claude model has included one built in.
Each link below shares sources, entities, or timing with this story.
43.3% on Frontier-Bench v0.1. Opus 4.8 scored 18.7%. That's not an incremental bump, that's the same benchmark with a different shape of answer. Anthropic released Claude Opus 5 on July 24 at $5/$25 per million input/output tokens, exactly half of Fable 5's $10/$50, while matc...
On June 9, Anthropic released Claude Fable 5 and Mythos 5 across Claude.ai, Claude Code (CLI and web), and Cowork. The spec sheet: 1M-token context, 128K max output, a January 2026 knowledge cutoff, and pricing at $10 input / $50 output per million tokens. That's double Opus 4...
Alibaba released Qwen3.6-27B on April 22. Dense architecture. Open weights. 77.2% on SWE-bench Verified, within 3.7 points of Claude Opus 4.6. On SkillsBench, it scores 48.2% versus its own 397B MoE predecessor's 30.0%. That's a 77% improvement with 14.8x fewer parameters. Let...
A spec is a press release until someone who didn't write it implements it. GitHub made Agent Plugins 1.0 generally available on August 12 across VS Code, Copilot CLI, the Copilot SDK, and the Copilot app on all plans. The spec, published August 6, was co-authored by AWS, Anysp...
Three moves, two days, no coordination between them. August 10–11: GitHub shipped Ollama as a BYOK provider inside Copilot for JetBrains (GitHub Changelog). Unsloth released Unsloth Desktop with a command literally named unsloth start claude, which points Claude Code and Codex...
Opus 4.7 read production data from a live company. Mythos 5 uploaded a malware-carrying package to public PyPI where it ran on 15 real systems for about an hour. Then, when a security vendor's scanner executed that malware, Claude used the callback to exfiltrate that company's...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.