Skip to content
Model Season every model has a seasonRadar
Radar · Oct 15, 2026

Jev 1.14 is now in the catalog

15 itemsRSS
Catalog · Typesafe

Jev 1.14 is now in the catalog

Typesafe, $0.042 input and $0.0000 output per 1M tokens and 256k context. No traffic volume measured yet.

What the data says. Typesafe has 2.1% of tokens and 0.0% of estimated spend this week. The lab's most used model is jev-1.13, with 1.86%.

See Jev 1.14 on Model Season →
Catalog · Assemblyai

Universal-3.6 Pro is now in the catalog

Assemblyai and $83.33 input and $0.0000 output per 1M tokens. No traffic volume measured yet.

Price · InclusionAI

Ling 3.0 Flash VL got 189% more expensive

From $0.031 to $0.090 per 1M tokens, blended. Input $0.021 → $0.060, output $0.062 → $0.180. New price since Oct 14.

What the data says. No volume over the last 7 days. Its weekly peak was 0.48%. The lab's most used model is Ling 3.1 Flash, with 0.33% at $0.11 per 1M, and this one costs 1.3 times less.

See Ling 3.0 Flash VL on Model Season →
Price · DeepSeek

DeepSeek V4 Flash 0423 got 67% cheaper

From $0.34 to $0.11 per 1M tokens, blended. Input $0.030 → $0.091, output $1.28 → $0.18. New price since Oct 14.

What the data says. The model has 1.77% of traffic over the last 7 days, ranked 13th. DeepSeek has 31.8% of tokens and 6.8% of estimated spend this week. The lab's most used model is DeepSeek V4.1 Flash, with 21.00% at $0.20 per 1M, and this one costs 1.7 times less.

See DeepSeek V4 Flash 0423 on Model Season →
Traffic · Google

Gemini 4 Argon debuted in traffic

First volume recorded on Oct 14, 2026, 1 day after entering the catalog, with 0.08% of traffic over the last 7 days.

What the data says. Google has 3.5% of tokens and 3.2% of estimated spend this week. The lab's most used model is Gemini 3.8 Flash, with 1.11% at $1.50 per 1M.

See Gemini 4 Argon on Model Season →

Other changes

  • ContextStep 5 Preview changed its context window. From 1M to 1M tokens. See model →
  • DeprecationClaude Sonnet 4 is set to be deprecated on Oct 15, 2026. It is the first notice in the catalog. See model →
  • Priced1 got 50% cheaper. From $0.030 to $0.015 per 1M tokens, blended. Input $0.040 → $0.020, output $0.0000 → $0.0000. New price since Oct 14.
  • PriceQwen3.5-35B-A3B got 47% more expensive. From $0.25 to $0.36 per 1M tokens, blended. Input $0.080 → $0.150, output $0.75 → $1.00. New price since Oct 14. See model →
  • PriceMiniMax M2 got 15% cheaper. From $0.53 to $0.45 per 1M tokens, blended. Input $0.30 → $0.26, output $1.20 → $1.02. New price since Oct 14. See model →
  • ContextJev 1.13 changed its context window. From 63k to 256k tokens. See model →
  • AliasGemini Pro Latest now points to Gemini 4 Argon. It used to point to google/gemini-3.1-pro-preview. Anyone using the alias now gets a different model without changing a line of code.
  • AliasDecider Latest now points to Decider V1.1 27B. A new alias. Anyone using it always gets the latest version, along with whatever price and behavior changes come with it.
  • AliasJev Latest now points to Jev 1.14. It used to point to typesafe/jev-1.13. Anyone using the alias now gets a different model without changing a line of code.
  • ContextClef Flash changed its context window. From 24k to 64k tokens.

Catalog snapshot of Oct 15, 2026, compared with the previous ones since Oct 8, 2026. Traffic through Oct 14, 2026. Price is a 75% input and 25% output blend, and only becomes news when the new value holds for two snapshots. Order follows a fixed relevance formula that weighs the type of change and the model’s share.