Token Ledger Digest – 2026-05-20
Lead change – biggest cost impact
-
Google Gemini Flash Latest (
~google/gemini-flash-latest)- Prompt price rose from $0.50/1M to $1.50/1M (+$1.00/1M).
- Completion price rose from $3.00/1M to $9.00/1M (+$6.00/1M).
- Who should care: Teams running high‑volume inference on this model will see per‑million‑token costs jump by $7.00; consider alternatives or prompt‑completion optimization.
Other price changes
-
Z.ai GLM 5.1 (
z-ai/glm-5.1)- Prompt price dropped from $0.98/1M to $0.00/1M.
- Completion price dropped from $3.08/1M to $0.00/1M.
- Who should care: Users can now run this model at zero token cost; ideal for cost‑sensitive prototypes or batch workloads.
-
Qwen: Qwen3.6 35B A3B (
qwen/qwen3.6-35b-a3b)- Prompt price fell slightly from $0.15/1M to $0.149/1M (‑$0.001/1M).
- Completion price unchanged at $1.00/1M.
- Who should care: Negligible impact; monitor for further drift.
-
Qwen: Qwen3.5‑35B‑A3B (
qwen/qwen3.5-35b-a3b)- Prompt price fell from $0.14/1M to $0.139/1M (‑$0.001/1M).
- Completion price unchanged at $1.00/1M.
- Who should care: Minimal effect; no action needed.
New model added
-
Google Gemini 3.5 Flash (
google/gemini-3.5-flash)- Prompt price: $1.50/1M.
- Completion price: $9.00/1M.
- Context window: 1,048,576 tokens.
- Who should care: Developers needing very long contexts; compare pricing against other long‑context options.
Summary
Total models tracked: 357. No other meaningful changes today.
Originally published at The Token Ledger. Subscribe for the daily digest.
