diff --git a/headroom/providers/anthropic.py b/headroom/providers/anthropic.py
index e1ff0126c..56a8d0790 100644
--- a/headroom/providers/anthropic.py
+++ b/headroom/providers/anthropic.py
@@ -83,12 +83,22 @@ def sanitize_anthropic_model_metadata(value: Any) -> Any:
 # Anthropic model context limits
 # All Claude 3+ models have 200K context
 ANTHROPIC_CONTEXT_LIMITS: dict[str, int] = {
+    # Claude Fable 5 - 1M context
+    "claude-fable-5": 1000000,
+    # Claude Opus 4.8 - 1M context
+    "claude-opus-4-8": 1000000,
     # Claude 4.7 (Opus 4.7) - 1M context
     "claude-opus-4-7": 1000000,
     # Claude 4.6 (Opus 4.6) - 1M context
     "claude-opus-4-6": 1000000,
     # Claude 4.5 (Opus 4.5)
     "claude-opus-4-5-20251101": 200000,
+    # Claude Sonnet 5 - 1M context
+    "claude-sonnet-5": 1000000,
+    # Claude Sonnet 4.6 - 1M context window
+    "claude-sonnet-4-6": 1000000,
+    # Claude Sonnet 4.5
+    "claude-sonnet-4-5": 200000,
     # Claude 4 (Sonnet 4, Haiku 4)
     "claude-sonnet-4-20250514": 200000,
     "claude-haiku-4-5-20251001": 200000,
@@ -110,17 +120,25 @@ def sanitize_anthropic_model_metadata(value: Any) -> Any:
 
 # Fallback pricing - LiteLLM is preferred source
 # NOTE: These are ESTIMATES. Always verify against actual Anthropic billing.
-# Last updated: 2025-01-14
+# Last updated: 2026-07-04
 ANTHROPIC_PRICING: dict[str, dict[str, float]] = {
-    # Claude 4.7 (Opus tier pricing)
-    "claude-opus-4-7": {"input": 15.00, "output": 75.00, "cached_input": 1.50},
-    # Claude 4.6 (Opus tier pricing)
-    "claude-opus-4-6": {"input": 15.00, "output": 75.00, "cached_input": 1.50},
-    # Claude 4.5 (Opus tier pricing)
-    "claude-opus-4-5-20251101": {"input": 15.00, "output": 75.00, "cached_input": 1.50},
+    # Claude Fable 5 (anthropic.com/pricing): $10 in / $50 out, cache read $1.
+    "claude-fable-5": {"input": 10.00, "output": 50.00, "cached_input": 1.00},
+    # Claude Opus 4.8 — current Opus tier: $5 in / $25 out, cache read $0.50.
+    "claude-opus-4-8": {"input": 5.00, "output": 25.00, "cached_input": 0.50},
+    # Claude 4.7 (current Opus tier)
+    "claude-opus-4-7": {"input": 5.00, "output": 25.00, "cached_input": 0.50},
+    # Claude 4.6 (current Opus tier)
+    "claude-opus-4-6": {"input": 5.00, "output": 25.00, "cached_input": 0.50},
+    # Claude 4.5 (current Opus tier — same rates as 4.6–4.8)
+    "claude-opus-4-5-20251101": {"input": 5.00, "output": 25.00, "cached_input": 0.50},
+    # Claude Sonnet 5 / 4.6 / 4.5 (current Sonnet tier): $3 in / $15 out, cache read $0.30
+    "claude-sonnet-5": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
+    "claude-sonnet-4-6": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
+    "claude-sonnet-4-5": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
     # Claude 4 (Sonnet/Haiku tier pricing)
     "claude-sonnet-4-20250514": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
-    "claude-haiku-4-5-20251001": {"input": 0.80, "output": 4.00, "cached_input": 0.08},
+    "claude-haiku-4-5-20251001": {"input": 1.00, "output": 5.00, "cached_input": 0.10},
     # Claude 3.5
     "claude-3-5-sonnet-20241022": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
     "claude-3-5-sonnet-latest": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
@@ -136,7 +154,7 @@ def sanitize_anthropic_model_metadata(value: Any) -> Any:
 # Default limits for pattern-based inference
 # Used when a model isn't in the explicit list but matches a known pattern
 _PATTERN_DEFAULTS = {
-    "opus": {"context": 200000, "pricing": {"input": 15.00, "output": 75.00, "cached_input": 1.50}},
+    "opus": {"context": 200000, "pricing": {"input": 5.00, "output": 25.00, "cached_input": 0.50}},
     "sonnet": {
         "context": 200000,
         "pricing": {"input": 3.00, "output": 15.00, "cached_input": 0.30},
diff --git a/tests/test_provider_model_fallback.py b/tests/test_provider_model_fallback.py
index 5911e12ee..9de7bc1e1 100644
--- a/tests/test_provider_model_fallback.py
+++ b/tests/test_provider_model_fallback.py
@@ -94,17 +94,17 @@ def test_pattern_based_inference_opus(self):
         assert limit == 200000
 
         pricing = provider._get_pricing("claude-opus-5-20260101")
-        assert pricing["input"] == 15.00
-        assert pricing["output"] == 75.00
+        assert pricing["input"] == 5.00
+        assert pricing["output"] == 25.00
 
     def test_pattern_based_inference_sonnet(self):
         """Test pattern-based inference for sonnet models."""
         provider = AnthropicProvider()
 
-        limit = provider.get_context_limit("claude-sonnet-5-20260101")
+        limit = provider.get_context_limit("claude-sonnet-6-20260101")
         assert limit == 200000
 
-        pricing = provider._get_pricing("claude-sonnet-5-20260101")
+        pricing = provider._get_pricing("claude-sonnet-6-20260101")
         assert pricing["input"] == 3.00
         assert pricing["output"] == 15.00
 
@@ -159,9 +159,9 @@ def test_pricing_for_known_models(self):
 
         # Claude Opus 4.5
         pricing = provider._get_pricing("claude-opus-4-5-20251101")
-        assert pricing["input"] == 15.00
-        assert pricing["output"] == 75.00
-        assert pricing["cached_input"] == 1.50
+        assert pricing["input"] == 5.00
+        assert pricing["output"] == 25.00
+        assert pricing["cached_input"] == 0.50
 
     def test_cost_estimation_for_new_models(self):
         """Test cost estimation works for new models."""
@@ -174,8 +174,8 @@ def test_cost_estimation_for_new_models(self):
             cached_tokens=0,
         )
 
-        # $15/1M input + $75/1M * 0.1M output = $15 + $7.5 = $22.5
-        assert cost == pytest.approx(22.5, rel=0.01)
+        # $5/1M input + $25/1M * 0.1M output = $5 + $2.5 = $7.5
+        assert cost == pytest.approx(7.5, rel=0.01)
 
 
 class TestAnthropicConfigLoading:
diff --git a/tests/test_providers/test_anthropic.py b/tests/test_providers/test_anthropic.py
index cdf503627..cbaf2accc 100644
--- a/tests/test_providers/test_anthropic.py
+++ b/tests/test_providers/test_anthropic.py
@@ -77,6 +77,11 @@ def test_get_context_limit_claude_opus(self, anthropic_provider):
     def test_get_context_limit_strips_ansi_model_suffix(self, anthropic_provider):
         assert anthropic_provider.get_context_limit("claude-opus-4-7[1m]") == 1000000
 
+    def test_get_context_limit_claude_5_family(self, anthropic_provider):
+        assert anthropic_provider.get_context_limit("claude-fable-5") == 1000000
+        assert anthropic_provider.get_context_limit("claude-opus-4-8") == 1000000
+        assert anthropic_provider.get_context_limit("claude-sonnet-5") == 1000000
+
     def test_supports_model_known(self, anthropic_provider):
         assert anthropic_provider.supports_model("claude-3-5-sonnet-20241022")
 
@@ -110,3 +115,13 @@ def test_pricing_lookup_strips_ansi_model_suffix(self, anthropic_provider):
         assert anthropic_provider._get_pricing("claude-opus-4-7[1m]") == (
             anthropic_provider._get_pricing("claude-opus-4-7")
         )
+
+    def test_pricing_claude_5_family(self, anthropic_provider):
+        fable = anthropic_provider._get_pricing("claude-fable-5")
+        assert fable == {"input": 10.00, "output": 50.00, "cached_input": 1.00}
+
+        opus = anthropic_provider._get_pricing("claude-opus-4-8")
+        assert opus == {"input": 5.00, "output": 25.00, "cached_input": 0.50}
+
+        sonnet = anthropic_provider._get_pricing("claude-sonnet-5")
+        assert sonnet == {"input": 3.00, "output": 15.00, "cached_input": 0.30}