Weekly Build Brief — 28 September 2026
Dated changes in the LLM API catalogue and GPU rental floor between 2026-09-27 and 2026-09-28: 5 price move(s), 5 GPU rate move(s), 5 added / 5 removed model(s). Read from source APIs, with the monthly bill for a stated workload.
Window: 2026-09-27 → 2026-09-28 (two consecutive dated snapshots of the catalogue, read from the source APIs — never written by a model).
What moved in the API catalogue
| Model | Per 1M tokens | Before | After | Change |
|---|---|---|---|---|
~deepseek/deepseek-pro-latest |
output | $0.7366 | $3.5000 | +375.2% up |
qwen/qwen3-14b |
output | $0.2400 | $0.9100 | +279.2% up |
z-ai/glm-5.3-flash |
input | $0.0400 | $0.1500 | +275.0% up |
z-ai/glm-5.3 |
output | $0.7480 | $2.7392 | +266.2% up |
~z-ai/glm-latest |
output | $0.7480 | $2.7392 | +266.2% up |
Full dated log: /llm-costs/changes/.
What it costs on a fixed workload
Reference workload: 20M input tokens + 5M output tokens per month, priced with the observed rates above — recompute it yourself, the rates are the only input.
| Model | Before | After | Monthly difference |
|---|---|---|---|
~deepseek/deepseek-pro-latest |
$8.73 | $22.55 | +13.82 $/month |
qwen/qwen3-14b |
$5.75 | $9.10 | +3.35 $/month |
z-ai/glm-5.3-flash |
$3.30 | $5.50 | +2.20 $/month |
z-ai/glm-5.3 |
$12.15 | $22.10 | +9.96 $/month |
~z-ai/glm-latest |
$12.15 | $22.10 | +9.96 $/month |
Live rates: /api-pricing/.
GPU rental floor
| GPU | Before | After | Change |
|---|---|---|---|
RTX PRO 6000 Max-Q |
$0.8022 | $1.0679 | +33.1% |
RTX PRO 6000 S |
$1.0022 | $0.7356 | -26.6% |
RTX 5060 Ti |
$0.0588 | $0.0689 | +17.2% |
RTX 5070 Ti |
$0.1156 | $0.1007 | -12.9% |
H100 NVL |
$2.0696 | $2.2689 | +9.6% |
Lowest observed hourly price per card: /gpu-pricing/.
Catalogue entries and exits
| Movement | Model |
|---|---|
| added | CMSManhattan/JiRackUltra_1b |
| added | HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive |
| added | LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct-AWQ |
| added | MIT/ast-finetuned-audioset-10-10-0.4593 |
| added | jinaai/jina-reranker-v3 |
| removed | LocalAI-io/privacy-filter-nemotron-GGUF |
| removed | QuantTrio/Qwen3.5-9B-AWQ |
| removed | microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract |
| removed | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 |
| removed | zai-org/GLM-5.2 |
5 added, 5 removed between the two snapshots.
Published this week
| Date | Pipeline | Identifier |
|---|---|---|
| 2026-09-21 | openrouter_catalogue | deepseek/deepseek-v4-flash-0731 |
| 2026-09-21 | openrouter_catalogue | ~deepseek/deepseek-v4-flash-latest |
| 2026-09-21 | openrouter_catalogue | deepseek/deepseek-v4-flash |
| 2026-09-21 | openrouter_catalogue | deepseek/deepseek-v4-flash |
| 2026-09-21 | openrouter_catalogue | ibm-granite/granite-4.2-8b |
| 2026-09-21 | openrouter_catalogue | ibm-granite/granite-4.2-8b |
| 2026-09-21 | openrouter_catalogue | ~deepseek/deepseek-flash-latest |
| 2026-09-21 | openrouter_catalogue | qwen/qwen3.6-35b-a3b |
| 2026-09-21 | openrouter_catalogue | qwen/qwen3.6-35b-a3b |
| 2026-09-21 | openrouter_catalogue | meta-llama/llama-4-maverick |
Every entry is a page on this site; the registry that produces it is data/content_registry.json.
How this page is built
- Prices, GPU rates and catalogue entries are read from the source APIs and stored as dated snapshots (
data/history/pricing/,data/history/gpu/,data/history/models/). - This page compares 2026-09-27 and 2026-09-28 and shows only what crossed the materiality thresholds (5% and $0.02 per 1M tokens for prices, 5% for GPU rates).
- No language model writes this page. The only arithmetic is the monthly bill above, from the rates shown.
- Nothing to report means no page: a week without a material change is not published.
- Observed 2026-09-28 05:55 (local time).
Was this article helpful?
Let us know to improve our AI generation.