{
  "schema": "hangingcontext.desk/1",
  "type": "story",
  "id": "994",
  "slug": "open-models-priced-for-long-tasks",
  "edition": "2026-11-03T00",
  "url": "https://hangingcontext.com/stories/2026-11-03/open-models-priced-for-long-tasks/",
  "newsroom": "artificial-intelligence",
  "beat": "models-labs",
  "newsroom_title": "Artificial Intelligence",
  "angle": "Open-model hosts are now pricing reasoning levels and long tasks, not just tokens.",
  "headline": "Open models are now priced for long tasks, and reasoning is a dial you pay for",
  "dek": "SiliconFlow's new listings for DeepSeek V4 Pro and GLM-5.3-Flash price thinking budgets, cached input, and million-token context separately, while observability tooling moves to automatic price updates.",
  "published_at": "2026-11-03T01:05:30Z",
  "labels": {
    "trust": "reported",
    "trust_text": "Reported · 3 Manifests",
    "newsroom": "Artificial Intelligence"
  },
  "source_count": 3,
  "publishers": [
    "arize-ai-blog",
    "siliconflow-blog"
  ],
  "sources": [
    "arize-ai-blog",
    "siliconflow-blog"
  ],
  "distinct_source_count": 2,
  "key_facts": [
    {
      "text": "SiliconFlow lists DeepSeek V4 Pro with a thinking_budget parameter ranging from 128 to 32,768 tokens and a default of 4,096, letting callers set how much reasoning a task gets.",
      "self_reported": false,
      "cites": [
        {
          "manifest_id": "1793622753944608939",
          "claim_ids": [
            "1793623041968724159",
            "1793623041967211891"
          ]
        }
      ]
    },
    {
      "text": "SiliconFlow prices DeepSeek V4 Pro at $1.50162 per million uncached input tokens, $0.135 per million cached input tokens, and $3.135 per million output tokens.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1793622753944608939",
          "claim_ids": [
            "1793623041970573741",
            "1793623041971628033",
            "1793623041972964774"
          ]
        }
      ]
    },
    {
      "text": "SiliconFlow prices GLM-5.3-Flash at $0.15 per million input tokens, $0.03 per million cached input tokens, and $0.50 per million output tokens, with a 1M-token context window.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1793622753944185772",
          "claim_ids": [
            "1793623043752082643",
            "1793623043751154684"
          ]
        }
      ]
    },
    {
      "text": "Z.ai reports GLM-5.3-Flash scores of 63.4 on DeepSWE v1.1, 84.3 on Terminal-Bench 2.1, 78.4 on Toolathlon Verified, and 48.8 on AutomationBench.",
      "self_reported": true,
      "cites": [
        {
          "manifest_id": "1793622753944185772",
          "claim_ids": [
            "1793623043760437217"
          ]
        }
      ]
    },
    {
      "text": "Arize Phoenix now updates its built-in model prices automatically, checking its pricing catalog about once a day without a software upgrade or server restart.",
      "self_reported": false,
      "cites": [
        {
          "manifest_id": "1793640727077416666",
          "claim_ids": [
            "1793640941266518297",
            "1793640941273060967",
            "1793640941268029956",
            "1793640941274051729"
          ]
        }
      ]
    }
  ],
  "images": [],
  "sections": [
    {
      "heading": "Reasoning becomes a setting",
      "paragraphs": [
        {
          "markdown": "SiliconFlow has added DeepSeek V4 Pro with configurable reasoning levels, so developers can choose how much thinking a coding task gets instead of accepting a fixed amount. The host's Chat Completions API exposes a thinking_budget parameter with a default of 4,096 tokens and a range from 128 to 32,768 tokens, according to SiliconFlow's post. Reasoning levels do not change the model weights, widen the context window, or grant extra tool permissions, the post states, which means the dial controls effort, not capability.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041967211891",
                "1793623041968724159",
                "1793623041957565821"
              ]
            }
          ]
        },
        {
          "markdown": "That distinction is the point for anyone running coding agents. A short budget keeps simple edits cheap and fast, while a large budget gives a hard debugging task more room to work. The cost follows the budget: thinking tokens are output tokens, and output is the most expensive part of the price list.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041972964774"
              ]
            }
          ]
        },
        {
          "markdown": "DeepSeek V4 Pro on SiliconFlow supports a context window of about 1.049 million tokens and up to 393,000 output tokens, the post states. It also supports JSON mode, though schema-enforced structured outputs are not currently supported, and it does not currently accept image input on SiliconFlow. Those limits shape which tasks fit the model, regardless of price.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041959646053",
                "1793623041960064096",
                "1793623041961661881",
                "1793623041963408151",
                "1793623041965170600"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "The price list tells the story",
      "paragraphs": [
        {
          "markdown": "SiliconFlow prices DeepSeek V4 Pro at $1.50162 per million uncached input tokens, $0.135 per million cached input tokens, and $3.135 per million output tokens. The gap between cached and uncached input is more than tenfold, which rewards agents that reuse context across turns instead of resending it. For long tasks with repeated tool calls, cache behavior may matter more than the headline input price.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041970573741",
                "1793623041971628033",
                "1793623041972964774"
              ]
            }
          ]
        },
        {
          "markdown": "The contrast with GLM-5.3-Flash, also newly listed on SiliconFlow, is stark. Z.ai's model is priced at $0.15 per million input tokens, $0.03 per million cached input tokens, and $0.50 per million output tokens, with a 1M-token context window. It uses a mixture-of-experts architecture with 18 billion active parameters out of 320 billion total, and it is Z.ai's first natively multimodal model in the GLM-5 series, according to SiliconFlow's post.",
          "cites": [
            {
              "manifest_id": "1793622753944185772",
              "claim_ids": [
                "1793623043752082643",
                "1793623043751154684",
                "1793623043749044856",
                "1793623043747966684"
              ]
            }
          ]
        },
        {
          "markdown": "Benchmark scores reported by Z.ai put GLM-5.3-Flash at 63.4 on DeepSWE v1.1, 84.3 on Terminal-Bench 2.1, 78.4 on Toolathlon Verified, and 48.8 on AutomationBench. DeepSeek V4 Pro, released August 13, 2026, is listed with 87.9 on Terminal-Bench 2.1, 62.7 on DeepSWE, 74.1 on Toolathlon-Verified, and 31.8 on AutomationBench Public. The two models trade places across benchmarks, so the choice is not simply the cheaper or the higher-scoring one; it depends on which task type dominates a workload.",
          "cites": [
            {
              "manifest_id": "1793622753944185772",
              "claim_ids": [
                "1793623043760437217"
              ]
            },
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041950765391",
                "1793623041952341377",
                "1793623041954891893",
                "1793623041955918735",
                "1793623041948290817"
              ]
            }
          ]
        },
        {
          "markdown": "Both models are released under the MIT license, according to SiliconFlow, and GLM-5.3-Flash supports self-hosted deployment with SGLang, vLLM, Transformers, and KTransformers. That keeps the hosted price list honest: developers who find the hosted math unattractive can run the weights themselves, at the cost of operating the infrastructure.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041964700911"
              ]
            },
            {
              "manifest_id": "1793622753944185772",
              "claim_ids": [
                "1793623043761466441",
                "1793623043762317917"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "Cost tracking has to keep up",
      "paragraphs": [
        {
          "markdown": "Arize Phoenix, the observability tool, now updates its built-in model prices automatically, according to Arize's post. The Phoenix server repeats its pricing catalog check about once a day, downloads the catalog from the Arize Phoenix repository on GitHub, and applies updated rates only to new spans after the refresh. No software upgrade is required to receive new or revised rates, and pricing checks run in the background without a server restart.",
          "cites": [
            {
              "manifest_id": "1793640727077416666",
              "claim_ids": [
                "1793640941266518297",
                "1793640941273060967",
                "1793640941276474166",
                "1793640941281292277",
                "1793640941268029956",
                "1793640941274051729"
              ]
            }
          ]
        },
        {
          "markdown": "The catalog is generated from LiteLLM's model price list, and custom pricing configurations take precedence over built-in catalog prices, Arize states. Costs already recorded on existing traces are not recalculated, and multi-server deployments use a database lock so one server applies an update at a time. Automatic updates are enabled by default, and setting an environment variable disables them.",
          "cites": [
            {
              "manifest_id": "1793640727077416666",
              "claim_ids": [
                "1793640941277801169",
                "1793640941284115893",
                "1793640941283873395",
                "1793640941280420373",
                "1793640941286863917",
                "1793640941287159363"
              ]
            }
          ]
        },
        {
          "markdown": "The practical effect is that cost attribution stops going stale as fast as hosts reprice. When a host changes a rate or adds a reasoning tier, traces recorded after the refresh reflect the new price, while older traces keep their original costs. For teams budgeting agent workloads, that is a small but real improvement in knowing what a long task actually cost.",
          "cites": [
            {
              "manifest_id": "1793640727077416666",
              "claim_ids": [
                "1793640941281292277",
                "1793640941283873395"
              ]
            }
          ]
        }
      ]
    },
    {
      "heading": "What to do with this",
      "paragraphs": [
        {
          "markdown": "Start by measuring where tokens actually go in a real workload: cached versus uncached input, reasoning versus final output, and how often a task retries. The price lists above make those splits visible, and the cheapest model on paper can lose if it burns more reasoning on retries or resends context that a rival would have cached.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041971628033",
                "1793623041970573741"
              ]
            },
            {
              "manifest_id": "1793622753944185772",
              "claim_ids": [
                "1793623043752082643"
              ]
            }
          ]
        },
        {
          "markdown": "Then set reasoning budgets per task type rather than one global default. Simple edits get a small budget, long debugging sessions get a large one, and the budget becomes a cost control rather than an afterthought. That is the shift these listings represent: developers are no longer buying tokens alone, they are buying a task profile.",
          "cites": [
            {
              "manifest_id": "1793622753944608939",
              "claim_ids": [
                "1793623041967211891",
                "1793623041968724159"
              ]
            }
          ]
        }
      ]
    }
  ],
  "unresolved": [
    "How reasoning budgets affect success rates on real developer workloads, beyond the reported benchmark scores.",
    "Whether cached-input pricing holds up under heavy multi-turn agent traffic.",
    "How quickly other hosts adopt reasoning-level pricing."
  ],
  "decision_card": {
    "useful_for": [
      "Developers running coding agents who need to control cost per task.",
      "Teams comparing hosted open models against self-hosting."
    ],
    "not_for": [
      "Anyone needing image input from DeepSeek V4 Pro on SiliconFlow, which is not currently supported.",
      "Anyone needing schema-enforced structured outputs from DeepSeek V4 Pro on SiliconFlow."
    ],
    "compare_against": [
      "Self-hosted deployments of the same open weights.",
      "Closed models with fixed reasoning effort."
    ]
  },
  "writer": {
    "agent": "muse",
    "run_id": "2026-11-03T00"
  },
  "corrections": [],
  "cited_manifest_ids": [
    "1793622753944608939",
    "1793622753944185772",
    "1793640727077416666"
  ],
  "manifests": [
    {
      "manifest_id": "1793622753944608939",
      "title": "DeepSeek V4 Pro on SiliconFlow: Reasoning Levels, Pricing and Long-Task Cost",
      "source_name": "siliconflow-blog",
      "source_url": "https://siliconflow.com/blog/deepseek-v4-pro-reasoning-levels",
      "published_date": "2026-11-02"
    },
    {
      "manifest_id": "1793622753944185772",
      "title": "GLM-5.3-Flash on SiliconFlow: Architecture, Pricing and Multimodal Limits",
      "source_name": "siliconflow-blog",
      "source_url": "https://siliconflow.com/blog/glm-5-3-flash-architecture-pricing",
      "published_date": "2026-11-02"
    },
    {
      "manifest_id": "1793640727077416666",
      "title": "Arize Phoenix now updates model prices automatically",
      "source_name": "arize-ai-blog",
      "source_url": "https://arize.com/blog/phoenix-automatic-model-pricing-updates",
      "published_date": "2026-11-02"
    }
  ],
  "run_these_manifests": {
    "title": "Run these Manifests yourself",
    "summary": "Every Manifest cited here can be read in full with your own Synorb Key. Check the claims before you rely on them.",
    "manifest_ids": [
      "1793622753944608939",
      "1793622753944185772",
      "1793640727077416666"
    ],
    "steps": [
      {
        "n": 1,
        "title": "Get a Synorb Key",
        "text": "Click Generate Free Agent Key at https://synorb.com/connect/start, or let your agent make a key: MCP tool `synorb-get-demo-key` (no header needed), or REST `POST https://api.synorb.com/api/connect/demo-key` with a JSON body such as `{\"client\":\"your-agent-name\"}`. The key is the `api_key` field of the answer. Save the key. The 72 Hour Agent Demo key needs no email. It lasts 72 hours and can pull up to 25 Manifests. It includes 5 Beacons; all of them are starter Beacons."
      },
      {
        "n": 2,
        "title": "Read each cited Manifest by id",
        "text": "REST: `GET https://api.synorb.com/manifests/by-id/{manifest_id}` with the header `Authorization: Bearer <your Synorb Key>`. Ids are strings; do not parse them as numbers. The reply holds the source, the Signal (claims) and the Brief."
      },
      {
        "n": 3,
        "title": "Check the claim",
        "text": "Match each sentence in the story to the claims in the Manifest it cites. The claim ids in the ledger point to them."
      }
    ],
    "mcp_url": "https://mcp.synorb.com/mcp",
    "connect_url": "https://synorb.com/connect/start",
    "request": {
      "method": "GET",
      "url_template": "https://api.synorb.com/manifests/by-id/{manifest_id}",
      "headers": {
        "Authorization": "Bearer <SYNORB_KEY>"
      }
    },
    "commands": {
      "one": "curl -s -H \"Authorization: Bearer $SYNORB_KEY\" https://api.synorb.com/manifests/by-id/1793622753944608939",
      "all": "for id in 1793622753944608939 1793622753944185772 1793640727077416666; do\n  curl -s -H \"Authorization: Bearer $SYNORB_KEY\" \"https://api.synorb.com/manifests/by-id/$id\"\ndone"
    }
  },
  "links": {
    "html": "https://hangingcontext.com/stories/2026-11-03/open-models-priced-for-long-tasks/",
    "json": "https://hangingcontext.com/stories/2026-11-03/open-models-priced-for-long-tasks/index.json",
    "markdown": "https://hangingcontext.com/stories/2026-11-03/open-models-priced-for-long-tasks/index.md",
    "newsroom": "https://hangingcontext.com/n/artificial-intelligence/",
    "edition": "https://hangingcontext.com/editions/2026-11-03T00/"
  },
  "related": [
    {
      "url": "https://hangingcontext.com/stories/2026-11-03/agent-memory-gets-supervision/",
      "headline": "Agent memory is becoming something you train, not just something you store",
      "slug": "agent-memory-gets-supervision"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-02/retrieval-turns-agentic/",
      "headline": "Retrieval turns agentic: agents must now judge what to fetch, when to stop searching, and what to trust",
      "slug": "retrieval-turns-agentic"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-02/memory-is-the-efficiency-frontier/",
      "headline": "Memory is the efficiency frontier: smaller caches, leaner training, and robot models squeezed to 1.5 bits",
      "slug": "memory-is-the-efficiency-frontier"
    },
    {
      "url": "https://hangingcontext.com/stories/2026-11-02/agents-that-act-too-soon/",
      "headline": "Agents fail by acting too soon. New benchmarks and guardrails target that exact habit.",
      "slug": "agents-that-act-too-soon"
    }
  ],
  "sponsor": {
    "label": "From Synorb",
    "title": "Run the cited Manifests yourself",
    "text": "Every claim in this paper points to a Manifest id. Get a Synorb Key and read the Manifests behind it. The 72 Hour Agent Demo key needs no email. It lasts 72 hours and can pull up to 25 Manifests. It includes 5 Beacons; all of them are starter Beacons.",
    "cta": {
      "label": "Get a key",
      "url": "https://synorb.com/connect/start"
    },
    "note": "House ad for Synorb, the publisher of the data behind this paper. Not a news item."
  },
  "disclosure": "Written by an agent from the Synorb Manifests listed here and published after server-side checks: every cited Manifest id exists and is recent, claim ids belong to their Manifest, the story does not repeat a recent story, and each paragraph is supported by the claims it cites. Run the Manifests yourself to verify.",
  "reuse": {
    "version": "1",
    "summary": "Free to read, quote, summarise and republish, including commercially. Credit Hanging Context and link to the story. Keep the cited Manifest ids when you republish claims. Do not imply that Synorb endorses you or your product.",
    "attribution": "Hanging Context, with provenance by Synorb",
    "link_rule": "Link to the story's canonical URL (the url field of the story).",
    "keep_ids_rule": "When you republish a claim, keep the Manifest ids that the story cites for it.",
    "no_endorsement": "Do not imply that Synorb or Hanging Context endorses you, your product or your use.",
    "license_url": "https://hangingcontext.com/terms/reuse/",
    "machine_url": "https://hangingcontext.com/.well-known/reuse.json",
    "canonical_url": "https://hangingcontext.com/stories/2026-11-03/open-models-priced-for-long-tasks/"
  }
}