Skip to content

Sync prices from the vendors' own sources - #1

Open
github-actions[bot] wants to merge 2 commits into
mainfrom
automation/sync
Open

github-actions[bot] wants to merge 2 commits into
mainfrom
automation/sync

Conversation

@github-actions

@github-actions github-actions Bot commented Sep 18, 2026

Copy link
Copy Markdown
Contributor

The scheduled run found changes in the vendors' published prices.

Each commit below is one entry, so every diff is the change to that
vendor alone. The last is the global.json version bump — the
published table is version-gated, so without it the app never sees any
of this.

Read it the way the checklist asks: in / out / cache_read against
the vendor's own page. A correct reading reports no change, so
anything here is a source disagreeing with what the entry carries.

Commits

  • ca6a457 Bump the version for 1 new
  • b10f118 Sync volcesark from its source — 1 new
Full run log

google-gemini  (https://aistudio.google.com/docs/pricing.md)
  ! 26 models with a Standard table, 7 of them in the entry

kimi  (https://platform.kimi.com/docs/pricing/chat.md)
  ! the vendor marks these retired: kimi-k2.5, moonshot-v1-8k, moonshot-v1-32k, moonshot-v1-128k, moonshot-v1-auto, moonshot-v1-8k-vision-preview, moonshot-v1-32k-vision-preview, moonshot-v1-128k-vision-preview, kimi-k2-0905-preview, kimi-k2-0711-preview, kimi-k2-turbo-preview, kimi-k2-thinking, kimi-k2-thinking-turbo

↯ stepfun  (https://platform.stepfun.com/docs/zh/guides/pricing/details.md + https://platform.stepfun.ai/docs/en/guides/pricing/details.md)
  https://platform.stepfun.com/docs/zh/guides/pricing/details.md: fetch failed
  Unreachable from this network, not read. The entry is unchanged.

stepfun-intl  (https://platform.stepfun.com/docs/zh/guides/pricing/details.md + https://platform.stepfun.ai/docs/en/guides/pricing/details.md)
  ! 3/3 rows priced from the vendor's international page

togetherai  (https://docs.together.ai/docs/serverless-models.md)
  ! 6/6 rows priced from the vendor's own docs table

openai  (https://developers.openai.com/api/docs/pricing.md)
  ! 37 priced models in the Standard table, 4 of them in the entry

openrouter  (https://openrouter.ai/api/v1/models + /{id}/endpoints)
  ! 17 of the entry's 20 models priced from their upstreams
  ! priced differently per upstream: deepseek/deepseek-v4.1-flash (22 endpoints, 11 prices), google/gemini-3.8-flash (6 endpoints, 3 prices), z-ai/glm-5.3-flash (30 endpoints, 9 prices), z-ai/glm-5.3 (34 endpoints, 18 prices), deepseek/deepseek-v4-flash-0731 (27 endpoints, 20 prices), anthropic/claude-opus-5 (11 endpoints, 3 prices), moonshotai/kimi-k3 (20 endpoints, 12 prices), openai/gpt-5.6-luna (7 endpoints, 4 prices), openai/gpt-5.6-sol (7 endpoints, 6 prices), tencent/hy3 (6 endpoints, 4 prices), anthropic/claude-sonnet-5 (10 endpoints, 2 prices), z-ai/glm-5.2 (30 endpoints, 17 prices), minimax/minimax-m3 (13 endpoints, 6 prices), deepseek/deepseek-v4-pro (16 endpoints, 13 prices), deepseek/deepseek-v4-flash (15 endpoints, 13 prices), xiaomi/mimo-v2.5 (6 endpoints, 5 prices)
  ! not resolved: deepseek-v4-flash-0731, deepseek-v4.1-flash, gemini-3.8-flash
  ! could not be read this run: deepseek/deepseek-v4.1-flash (no agreed price among 22 upstreams — left as is), google/gemini-3.8-flash (no agreed price among 6 upstreams — left as is), deepseek/deepseek-v4-flash-0731 (no agreed price among 27 upstreams — left as is)

qianwenai  (https://help.aliyun.com/zh/model-studio/billing-for-model-studio)
  ! cache rates are not read: these models are excepted from every published percentage (20% implicit, 10% explicit-hit) and the vendor points at the console instead

volcesark  (https://www.volcengine.com/api/doc/getDocDetail?DocumentID=1544106, https://www.volcengine.com/api/doc/getDocDetail?DocumentID=1925114, https://www.volcengine.com/api/doc/getDocDetail?DocumentID=2366394)
  + kimi-k2.8-preview is new

1 entr(ies) to change: 1 new
1 entr(ies) unreachable from this network — see the ↯ above; nothing is wrong with their pages
no source, still hand-maintained: baidu-qianfan-token-plan, kimi-for-coding, qianwenai-token-plan, tencent-token-plan
✓ wrote entries/volcesark/models.json
✓ bumped global.json version 48 -> 49
✓ committed entries/volcesark/models.json
✓ committed global.json

commits are local; nothing was pushed

Opened by the sync workflow at the vendors' own sources. Nothing
here is merged automatically.

github-actions Bot and others added 2 commits September 20, 2026 11:48
The published table is version-gated, so a price that moved without this
is a price the app never sees.

Co-Authored-By: Claude Code <noreply@anthropic.com>
landoyjx pushed a commit that referenced this pull request Sep 21, 2026
The entry's twenty rows were chosen by a weekly slice of OpenRouter's own usage
ranking, which quietly excludes anything that had a strong month but a quiet
latest week. The ranking script now takes `--days N` — aggregating daily rows
over the window, on `period=day`, because the week/month periods are sampled
server-side (a 4-month month query returns ~5 days of rows) — and the entry is
rewritten from a 90-day window: 77 resolved models (57 new), from the 100 ranked
(`--top 100`), out of 126 the window saw.

What the wider window changes is visible in the top ten: mimo-v2.5 is #1 (85.6T
tokens), the two deepseek-v4-flash snapshots are #2/#3, and the flagship moves
gpt-5.6-luna → mimo-v2.5, which is the ranking's answer to "which model is this
provider known for" — a better question than the price heuristic for an
aggregator. Rank #12 `stealth/ox-alpha` and a few `:free` mirrors are not in the
models API and stay unwritten, listed by the script; embeddings never join.

The models.dev-style price column still comes from OpenRouter's own models API
(an aggregate — this entry is aggregate by tag, so a third-party price is its
nature), and the next `fetch-all` run will re-price these rows to the modal
endpoint price, which is the sturdier read.

142 price rows, v49. 46 cross-provider price divergences are now published —
the point of carrying an aggregate at all.

Co-Authored-By: Claude Code <noreply@anthropic.com>

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants