DEV Community

4663437Mehdi
4663437Mehdi

Posted on • Originally published at 4663437mehdi.github.io

The Token Ledger Digest – 2026-08-03

The Token Ledger Digest – 2026-08-03

Most cost‑impacting change: Z.ai’s GLM 5.2 saw a sharp price rise, raising both prompt and completion costs by roughly $0.90‑$2.85 per 1M tokens.

Price changes

Model What changed Old price (/1M) New price (/1M) Who should care
Z.ai: GLM 5.2 Prompt ↑ from $0.28 to $1.19; Completion ↑ from $0.89 to $3.74 $0.28 / $0.89 $1.19 / $3.74 Teams using GLM 5.2 for high‑volume generation; budget reassessment needed.
Qwen: Qwen3.5‑122B‑A10B Prompt ↑ $0.26 → $0.40; Completion ↑ $2.08 → $3.20 $0.26 / $2.08 $0.40 / $3.20 Developers balancing quality vs cost for mid‑size LLMs.
Qwen: Qwen3 VL 235B A22B Thinking Prompt ↑ $0.40 → $0.98; Completion ↓ $4.00 → $3.95 $0.40 / $4.00 $0.98 / $3.95 Vision‑language users; prompt cost up slightly, completion marginally cheaper.
Qwen: Qwen3 235B A22B Instruct 2507 Prompt ↑ $0.09 → $0.15; Completion ↑ $0.55 → $0.60 $0.09 / $0.55 $0.15 / $0.60 Cost‑sensitive applications using this instruct variant.
OpenAI: GPT‑5.6 Luna Pro No meaningful change (prompt $0.10, completion $0.60 both unchanged) $0.10 / $0.60 $0.10 / $0.60 No action required.
OpenAI: GPT‑5.6 Luna Same as above $0.10 / $0.60 $0.10 / $0.60 No action required.
OpenAI: GPT‑5.6 Terra Pro No meaningful change (prompt $1.00, completion $6.00) $1.00 / $6.00 $1.00 / $6.00 No action required.
OpenAI: GPT‑5.6 Terra Same as above $1.00 / $6.00 $1.00 / $6.00 No action required.

Per‑token prices were multiplied by 1,000,000 and rounded to two decimal places for readability.

Cheapest models today (per‑million tokens)

  1. inclusionAI: Ling‑2.6‑flash – Prompt $0.01, Completion $0.03
  2. Mistral: Mistral Nemo – Prompt $0.02, Completion $0.03
  3. IBM: Granite 4.0 Micro – Prompt $0.02, Completion $0.11

No models were added or removed today. Total models tracked: 337.


Originally published at The Token Ledger. Subscribe for the daily digest.

Top comments (0)