llm.rb 16.0.0.pre.4 → 16.0.0.pre.6

data/data/openrouter.json

@@ -518,9 +518,9 @@
         "output": 943718
       },
       "cost": {
-        "input": 0.4,
-        "output": 10,
-        "cache_read": 0.4
+        "input": 0.29,
+        "output": 9,
+        "cache_read": 0.28
       }
     },
     "poolside/laguna-s-2.1:free": {
@@ -1859,13 +1859,13 @@
       },
       "open_weights": true,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 943718
       },
       "cost": {
-        "input": 0.018,
-        "output": 0.32,
-        "cache_read": 0.018
+        "input": 0.01,
+        "output": 1.28,
+        "cache_read": 0.01
       },
       "canonical_model_id": "deepseek/deepseek-v4-flash-0731"
     },
@@ -1910,9 +1910,9 @@
         "output": 943718
       },
       "cost": {
-        "input": 0.3,
-        "output": 1.2,
-        "cache_read": 0.006
+        "input": 0.0198,
+        "output": 0.396,
+        "cache_read": 0.00291
       },
       "canonical_model_id": "deepseek/deepseek-v4.1-flash"
     },
@@ -2067,12 +2067,12 @@
       "open_weights": true,
       "limit": {
         "context": 1048576,
-        "output": 943718
+        "output": 393216
       },
       "cost": {
-        "input": 0.4,
-        "output": 4.2,
-        "cache_read": 0.32
+        "input": 0.66,
+        "output": 1.98,
+        "cache_read": 0.022
       },
       "canonical_model_id": "deepseek/deepseek-v4-pro-0813"
     },
@@ -2219,9 +2219,9 @@
         "output": 384000
       },
       "cost": {
-        "input": 0.95526,
-        "output": 1.91052,
-        "cache_read": 0.079605
+        "input": 0.783,
+        "output": 1.566,
+        "cache_read": 0.06525
       },
       "canonical_model_id": "deepseek/deepseek-v4-pro"
     },
@@ -2335,12 +2335,12 @@
       "open_weights": true,
       "limit": {
         "context": 1048576,
-        "output": 131072
+        "output": 384000
       },
       "cost": {
-        "input": 0.14,
-        "output": 0.28,
-        "cache_read": 0.028
+        "input": 0.07854,
+        "output": 0.15708,
+        "cache_read": 0.015708
       },
       "canonical_model_id": "deepseek/deepseek-v4-flash"
     },
@@ -2551,9 +2551,9 @@
         "output": 128000
       },
       "cost": {
-        "input": 0.132,
-        "output": 0.528,
-        "cache_read": 0.033
+        "input": 0.0825,
+        "output": 0.33,
+        "cache_read": 0.020625
       },
       "canonical_model_id": "tencent/hy3"
     },
@@ -2623,9 +2623,9 @@
         "output": 64000
       },
       "cost": {
-        "input": 0.834,
-        "output": 2.501,
-        "cache_read": 0.042
+        "input": 0.7506,
+        "output": 2.2509,
+        "cache_read": 0.0378
       },
       "canonical_model_id": "tencent/hy4-preview"
     },
@@ -3142,7 +3142,7 @@
       "cost": {
         "input": 0.3,
         "output": 0.9,
-        "cache_read": 0.05
+        "cache_read": 0.055
       },
       "canonical_model_id": "zhipuai/glm-4.6v"
     },
@@ -3298,7 +3298,7 @@
       },
       "open_weights": true,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 943717
       },
       "cost": {
@@ -3590,9 +3590,9 @@
         "output": 943718
       },
       "cost": {
-        "input": 0.3249,
-        "output": 4.4,
-        "cache_read": 0.2599
+        "input": 0.432,
+        "output": 3.99,
+        "cache_read": 0.26
       },
       "canonical_model_id": "zhipuai/glm-5.2"
     },
@@ -3630,9 +3630,9 @@
         "output": 131072
       },
       "cost": {
-        "input": 1.4,
-        "output": 4.4,
-        "cache_read": 0.26
+        "input": 0.9646,
+        "output": 3.0316,
+        "cache_read": 0.17914
       },
       "canonical_model_id": "zhipuai/glm-5.1"
     },
@@ -3746,7 +3746,7 @@
       },
       "open_weights": true,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 943717
       },
       "cost": {
@@ -4430,11 +4430,11 @@
       "open_weights": true,
       "limit": {
         "context": 131072,
-        "output": 16384
+        "output": 117964
       },
       "cost": {
-        "input": 0.3,
-        "output": 1.2,
+        "input": 0.35,
+        "output": 1.5,
         "cache_read": 0.04
       },
       "canonical_model_id": "meta/muse-glimmer-30b"
@@ -4758,12 +4758,12 @@
       },
       "open_weights": false,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 943718
       },
       "cost": {
         "input": 0.02,
-        "output": 0.3,
+        "output": 0.2475,
         "cache_read": 0.01
       }
     },
@@ -4799,13 +4799,13 @@
       },
       "open_weights": false,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 131072
       },
       "cost": {
-        "input": 0.19,
+        "input": 0.13,
         "output": 4,
-        "cache_read": 0.19
+        "cache_read": 0.13
       }
     },
     "arcee-ai/trinity-large-thinking": {
@@ -4984,7 +4984,6 @@
         {
           "type": "effort",
           "values": [
-            "none",
             "low",
             "medium",
             "high",
@@ -5017,13 +5016,13 @@
       "cost": {
         "input": 2,
         "output": 10,
-        "cache_read": 0.2,
+        "cache_read": 0.1,
         "cache_write": 2.5,
         "tiers": [
           {
             "input": 4,
             "output": 15,
-            "cache_read": 0.4,
+            "cache_read": 0.2,
             "cache_write": 5,
             "tier": {
               "type": "context",
@@ -5034,7 +5033,7 @@
         "context_over_200k": {
           "input": 4,
           "output": 15,
-          "cache_read": 0.4,
+          "cache_read": 0.2,
           "cache_write": 5
         }
       }
@@ -6072,9 +6071,9 @@
         "output": 943718
       },
       "cost": {
-        "input": 0.02,
-        "output": 0.6,
-        "cache_read": 0.02
+        "input": 0.0198,
+        "output": 0.396,
+        "cache_read": 0.00291
       }
     },
     "~deepseek/deepseek-pro-latest": {
@@ -6113,12 +6112,12 @@
       "open_weights": false,
       "limit": {
         "context": 1048576,
-        "output": 943718
+        "output": 393216
       },
       "cost": {
-        "input": 0.15808,
-        "output": 1.9584,
-        "cache_read": 0.07488
+        "input": 0.123,
+        "output": 3.5,
+        "cache_read": 0.123
       }
     },
     "~deepseek/deepseek-v4-flash-latest": {
@@ -6156,13 +6155,13 @@
       },
       "open_weights": false,
       "limit": {
-        "context": 1310720,
+        "context": 1048576,
         "output": 943718
       },
       "cost": {
-        "input": 0.012,
-        "output": 0.7,
-        "cache_read": 0.01
+        "input": 0.0099,
+        "output": 0.13068,
+        "cache_read": 0.001386
       }
     },
     "~x-ai/grok-latest": {
@@ -6666,9 +6665,9 @@
         "output": 235929
       },
       "cost": {
-        "input": 0.0765,
-        "output": 0.255,
-        "cache_read": 0.0425
+        "input": 0.09,
+        "output": 0.3,
+        "cache_read": 0.05
       },
       "canonical_model_id": "google/gemma-4-26b-a4b-it"
     },
@@ -9041,8 +9040,8 @@
         "output": 235929
       },
       "cost": {
-        "input": 0.6562,
-        "output": 3.3,
+        "input": 0.6712,
+        "output": 3.35,
         "cache_read": 0.18
       },
       "canonical_model_id": "moonshotai/kimi-k2.7-code"
@@ -9212,7 +9211,7 @@
       },
       "open_weights": true,
       "limit": {
-        "context": 1000000,
+        "context": 262144,
         "output": 32768
       },
       "cost": {
@@ -9717,7 +9716,7 @@
       },
       "open_weights": true,
       "limit": {
-        "context": 1048576,
+        "context": 1050000,
         "output": 131072
       },
       "cost": {
@@ -10865,9 +10864,9 @@
         "output": 235929
       },
       "cost": {
-        "input": 0.0647,
-        "output": 4.4,
-        "cache_read": 0.0517
+        "input": 0.0249,
+        "output": 4.35,
+        "cache_read": 0.0199
       },
       "canonical_model_id": "alibaba/qwen3.8-27b"
     },
@@ -11500,7 +11499,7 @@
       "open_weights": true,
       "limit": {
         "context": 262144,
-        "output": 235929
+        "output": 65536
       },
       "cost": {
         "input": 0.26,
@@ -12242,11 +12241,11 @@
       "open_weights": true,
       "limit": {
         "context": 131072,
-        "output": 8192
+        "output": 16384
       },
       "cost": {
-        "input": 0.13,
-        "output": 0.52
+        "input": 0.12,
+        "output": 0.5
       },
       "canonical_model_id": "alibaba/qwen3-30b-a3b"
     },
@@ -14148,6 +14147,73 @@
         "output": 2
       }
     },
+    "openai/gpt-6.1-sol": {
+      "id": "openai/gpt-6.1-sol",
+      "name": "GPT-6.1 Sol",
+      "description": "Near-Astra performance for complex coding, computer use, and professional work at a lower cost.",
+      "family": "gpt-sol",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [
+        {
+          "type": "effort",
+          "values": [
+            "low",
+            "medium",
+            "high",
+            "xhigh",
+            "max"
+          ]
+        }
+      ],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": false,
+      "knowledge": "2026-04-30",
+      "release_date": "2026-09-29",
+      "last_updated": "2026-09-29",
+      "modalities": {
+        "input": [
+          "text",
+          "image",
+          "pdf"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": false,
+      "limit": {
+        "context": 1050000,
+        "input": 922000,
+        "output": 128000
+      },
+      "cost": {
+        "input": 2,
+        "output": 10,
+        "cache_read": 0.1,
+        "cache_write": 2.5,
+        "tiers": [
+          {
+            "input": 4,
+            "output": 15,
+            "cache_read": 0.2,
+            "cache_write": 5,
+            "tier": {
+              "type": "context",
+              "size": 272000
+            }
+          }
+        ],
+        "context_over_200k": {
+          "input": 4,
+          "output": 15,
+          "cache_read": 0.2,
+          "cache_write": 5
+        }
+      },
+      "canonical_model_id": "openai/gpt-6.1-sol"
+    },
     "openai/gpt-4o-2024-08-06": {
       "id": "openai/gpt-4o-2024-08-06",
       "name": "GPT-4o (2024-08-06)",
@@ -15311,6 +15377,70 @@
       },
       "canonical_model_id": "openai/gpt-6-luna"
     },
+    "openai/gpt-6.1-sol-pro": {
+      "id": "openai/gpt-6.1-sol-pro",
+      "name": "GPT-6.1 Sol Pro",
+      "description": "Frontier GPT model for professional reasoning, coding, and multimodal work",
+      "family": "gpt",
+      "attachment": true,
+      "reasoning": true,
+      "reasoning_options": [
+        {
+          "type": "effort",
+          "values": [
+            "low",
+            "medium",
+            "high",
+            "xhigh",
+            "max"
+          ]
+        }
+      ],
+      "tool_call": true,
+      "structured_output": true,
+      "temperature": false,
+      "release_date": "2026-09-29",
+      "last_updated": "2026-09-29",
+      "modalities": {
+        "input": [
+          "pdf",
+          "image",
+          "text"
+        ],
+        "output": [
+          "text"
+        ]
+      },
+      "open_weights": false,
+      "limit": {
+        "context": 1050000,
+        "output": 128000
+      },
+      "cost": {
+        "input": 2,
+        "output": 10,
+        "cache_read": 0.1,
+        "cache_write": 2.5,
+        "tiers": [
+          {
+            "input": 4,
+            "output": 15,
+            "cache_read": 0.2,
+            "cache_write": 5,
+            "tier": {
+              "type": "context",
+              "size": 272000
+            }
+          }
+        ],
+        "context_over_200k": {
+          "input": 4,
+          "output": 15,
+          "cache_read": 0.2,
+          "cache_write": 5
+        }
+      }
+    },
     "openai/o4-mini-high": {
       "id": "openai/o4-mini-high",
       "name": "o4 Mini High",
@@ -15552,12 +15682,11 @@
       "open_weights": true,
       "limit": {
         "context": 131072,
-        "output": 65536
+        "output": 117964
       },
       "cost": {
-        "input": 0.15,
-        "output": 0.6,
-        "cache_read": 0.075
+        "input": 0.037,
+        "output": 0.17
       },
       "canonical_model_id": "openai/gpt-oss-120b"
     },

data/docs/deepdive/advanced/context.md

@@ -424,7 +424,9 @@ The block receives the prompt object when it takes an argument, and
 otherwise runs in the prompt's context:
 [`LLM::Prompt#to_a`](https://r.uby.dev/api-docs/llm.rb/LLM/Prompt.html#to_a)
 returns the messages in order, and two prompts are equal when their
-messages match.
+messages match. Each of `system`, `user`, `developer`, and `talk`
+accepts an `extra:` hash, which is attached to the message it builds,
+so a caller can carry a field of its own alongside the content.
 
 #### Why would I use it?
 

data/docs/deepdive/features/console.md

@@ -66,9 +66,9 @@ should use the `console` names.
 When `bin/llm.rb` starts, it looks for an `AGENTS.md` file in the
 current working directory. If one exists, its contents become the
 agent's instructions for the session, so a project can describe how it
-wants the agent to work without passing flags. The instructions are
-injected once, so a resumed session that already has a system message
-keeps the one it has.
+wants the agent to work without passing flags. The agent brings its
+instructions up to date before each request, so a resumed session runs
+on the current file rather than the one it was saved with.
 
 The user-message label is exposed through
 [`LLM::Console#sender`](https://r.uby.dev/api-docs/llm.rb/LLM/Console.html#sender-instance_method),

data/docs/deepdive/features/database.md

@@ -34,13 +34,23 @@ and
 for serialization and
 [`LLM::Context#restore`](https://r.uby.dev/api-docs/llm.rb/LLM/Context.html#restore)
 for deserialization. Save writes the current state, restore loads
-it back and picks up where the conversation left off. The ORM
-wrappers automate this; each
-[`LLM::Agent#talk`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#talk)
-or
-[`LLM::Agent#ask`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#ask)
-call persists the
-updated state back to the column automatically.
+it back and picks up where the conversation left off.
+
+The ORM wrappers write on a different boundary than the file does, and
+the two are worth telling apart.
+[`LLM::Step`](https://r.uby.dev/api-docs/llm.rb/LLM/Step.html) is
+prepended onto the stream, and it saves the conversation through the
+record the context is bound to each time
+[`LLM::Stream#on_step`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html#on_step-instance_method)
+fires - after a response is in the conversation and before the tools it
+asked for run. A turn that runs tools before it answers is therefore
+saved more than once, and an application that writes it down can
+continue an interrupted turn instead of redoing it.
+
+A `path:` has no record behind it to write to, so that one saves once
+per turn. The difference is what each is for: the column is a
+checkpoint after every completed request, and the file is a session at
+the end of one.
 
 #### Why would I use it?
 
@@ -242,12 +252,10 @@ first call, a fresh agent is created and the conversation starts
 from scratch.
 
 On subsequent calls, the stored state is restored and the
-conversation continues. Every
-[`LLM::Agent#talk`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#talk)
-or
-[`LLM::Agent#ask`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#ask)
-call persists the
-updated state back to the column automatically.
+conversation continues. The column is written after each completed
+request rather than at the end of a turn, so a turn that runs tools
+before it answers is saved more than once - and a turn that is
+interrupted can be resumed from its last completed request.
 
 The `data_column:` option lets you use a different column name.
 The `format:` option controls the storage type. Use `:string` for
@@ -327,7 +335,7 @@ agent.talk "perform research"
 #### Why would I use it?
 
 The ActiveRecord wrapper integrates with your existing models.
-Agent state is automatically persisted after each conversation turn
+Agent state is automatically persisted after each completed request,
 using the same `create!`, `save!`, and query methods you already
 use.
 
@@ -447,11 +455,8 @@ the plugin in the model class. The `data` column stores
 the full agent state as JSON, same structure as ActiveRecord.
 On first call, a fresh agent is created. On subsequent calls,
 the stored state is restored and the conversation continues.
-Every
-[`LLM::Agent#talk`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#talk)
-or
-[`LLM::Agent#ask`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#ask)
-call persists the updated state.
+The state is written after each completed request, as it is under
+ActiveRecord.
 
 The `data_column:` and `format:` options work identically to
 ActiveRecord. For `:jsonb`, Sequel loads the `pg_json` extension

data/docs/deepdive/fundamentals/agents.md

@@ -21,8 +21,9 @@ the model responds, and if it requests tools the agent
 executes them automatically and feeds the results back. The tool
 loop can be bounded with
 [`LLM::Agent.tool_budget`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html#tool_budget-class_method)
-(see the Tool budget section). Instructions are injected once
-unless a system message is already present.
+(see the Tool budget section). Instructions are injected once and
+brought up to date from the agent before each request, unless the
+prompt already carries a system message of the caller's own.
 
 #### Why would I use it?
 

data/docs/deepdive/fundamentals/stream.md

@@ -67,6 +67,17 @@ receives tokens as they arrive.
 fires when the model requests a tool.
 [`LLM::Stream#on_tool_return`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html#on_tool_return)
 fires when the tool completes.
+[`LLM::Stream#on_step`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html#on_step-instance_method)
+fires when a request completes, which is the boundary between one
+request and the next: the response is in the conversation, and a
+turn that asked for tools runs them after this while a turn that is
+finished ends here. That boundary is where a conversation is whole,
+and the runtime's
+[`LLM::Step`](https://r.uby.dev/api-docs/llm.rb/LLM/Step.html) is
+written against it - it is prepended onto a stream, and saves the
+conversation through the record the context is bound to, so a turn
+that is interrupted can be continued from its last completed request
+rather than started over.
 [`LLM::Stream#on_retry`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html#on_retry)
 fires each time a failed request is retried. Compaction hooks
 let you show progress or log what was trimmed. Skill hooks bracket a
@@ -96,6 +107,11 @@ class MyStream < LLM::Stream
   def on_tool_return(tool, result)
   end
 
+  # A request has completed: the response is in the conversation, and
+  # any tools it asked for run after this.
+  def on_step(ctx, res)
+  end
+
   # Before a transformer rewrites an outgoing message.
   def on_transform(transformer)
   end
@@ -135,9 +151,3 @@ agent.talk "Explain Ruby fibers."
 A stream subclass gives you visibility into more than just content chunks. React to tool
 calls as they happen, show compaction progress, or integrate with
 an existing observability stack.
-
-#### Notes
-
-The IO-like form is equivalent to
-[`LLM::Stream#on_content`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html#on_content)
-and does not include the other hooks.