Skip to content

Commit c923de6

Browse files
authored
feat(providers): add Claude Fable 5 model (#5334)
Re-add claude-fable-5 (Anthropic's most capable widely released model), not marked recommended. Verified against live Anthropic API + docs: - Pricing $10/$50 per MTok, $1 cached input (cache-read); 1M context; 128k max output; GA 2026-06-09 - Adaptive thinking only (always on, cannot be disabled) — routed via supportsAdaptiveThinking; manual budget_tokens/temperature are rejected with 400 - effort levels low/medium/high/xhigh/max (default high) - nativeStructuredOutputs confirmed supported (live output_format json_schema call returned 200), matching same-generation siblings (Sonnet 5, Opus 4.8)
1 parent 9204b4a commit c923de6

2 files changed

Lines changed: 22 additions & 1 deletion

File tree

apps/sim/providers/anthropic/core.ts

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -84,6 +84,7 @@ const THINKING_BUDGET_TOKENS: Record<string, number> = {
8484

8585
/**
8686
* Checks if a model supports adaptive thinking (thinking.type: "adaptive").
87+
* Fable 5 supports ONLY adaptive thinking (always on; type: "disabled" is rejected).
8788
* Sonnet 5 supports ONLY adaptive thinking (manual budget_tokens returns a 400 error).
8889
* Opus 4.8 and Opus 4.7 support ONLY adaptive thinking (no extended thinking / budget_tokens).
8990
* Opus 4.6 and Sonnet 4.6 support both extended and adaptive thinking — use adaptive.
@@ -92,6 +93,7 @@ const THINKING_BUDGET_TOKENS: Record<string, number> = {
9293
function supportsAdaptiveThinking(modelId: string): boolean {
9394
const normalizedModel = modelId.toLowerCase()
9495
return (
96+
normalizedModel.includes('fable-5') ||
9597
normalizedModel.includes('sonnet-5') ||
9698
normalizedModel.includes('opus-4-8') ||
9799
normalizedModel.includes('opus-4.8') ||
@@ -107,7 +109,7 @@ function supportsAdaptiveThinking(modelId: string): boolean {
107109
/**
108110
* Builds the thinking configuration for the Anthropic API based on model capabilities and level.
109111
*
110-
* - Sonnet 5, Opus 4.8, Opus 4.7: Uses adaptive thinking only (no extended thinking support)
112+
* - Fable 5, Sonnet 5, Opus 4.8, Opus 4.7: Uses adaptive thinking only (no extended thinking support)
111113
* - Opus 4.6, Sonnet 4.6: Uses adaptive thinking with effort parameter
112114
* - Other models: Uses budget_tokens-based extended thinking
113115
*

apps/sim/providers/models.ts

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -669,6 +669,25 @@ export const PROVIDER_DEFINITIONS: Record<string, ProviderDefinition> = {
669669
toolUsageControl: true,
670670
},
671671
models: [
672+
{
673+
id: 'claude-fable-5',
674+
pricing: {
675+
input: 10.0,
676+
cachedInput: 1.0,
677+
output: 50.0,
678+
updatedAt: '2026-07-01',
679+
},
680+
capabilities: {
681+
nativeStructuredOutputs: true,
682+
maxOutputTokens: 128000,
683+
thinking: {
684+
levels: ['low', 'medium', 'high', 'xhigh', 'max'],
685+
default: 'high',
686+
},
687+
},
688+
contextWindow: 1000000,
689+
releaseDate: '2026-06-09',
690+
},
672691
{
673692
id: 'claude-sonnet-5',
674693
pricing: {

0 commit comments

Comments
 (0)