Launch ·
DeepSeek publishes DeepSeek-V4.1-Flash-1017 weights on Hugging Face
DeepSeek released DeepSeek-V4.1-Flash-1017, an image-text-to-text model, according to Hugging Face. The model is also listed on OpenRouter and Artificial Analysis.
DeepSeek publishes DeepSeek-V4.1-Flash-1017 weights on Hugging Face
News × data
The story published out there and, below it, what real token traffic from the last 7 days says about who is in it. The headline opens the original article; the model name opens its page here.
Pricing ·
DeepSeek prices search and fetch at $0.75 per 1000 successful requests
According to the DeepSeek API docs, the search and fetch tools cost $0.75 per 1000 successful requests. The Search API finds web pages for a query, and the Fetch API returns the content of a single page as Markdown.
Capability ·
Claude uncovers the origin of NIST curve seeds used in ECDSA
According to Anthropic, Claude discovered in under half an hour where the seeds of the NIST P-192, P-224, P-256, P-384 and P-521 curves came from. They were not hash outputs but came from the lrand48 generator seeded with a Unix timestamp. The company says this is not a cryptographic break or a weakness in the curves.
Changes in the data
Clef Flash got 81% more expensive
From $0.016 to $0.029 per 1M tokens, blended. Input $0.021 → $0.038, output $0.0000 → $0.0000. New price since Oct 18.
R1 was removed from the catalog
Last listed on Oct 18, 2026. Anyone calling this identifier needs to migrate.
What the data says. No volume over the last 7 days. Its weekly peak was 6.49%. DeepSeek has 30.1% of tokens and 3.3% of estimated spend this week. The lab's most used model is DeepSeek V4.1 Flash, with 20.01% at $0.046 per 1M.
See R1 on Model Season →WizardLM-2 8x22B was removed from the catalog
Last listed on Oct 18, 2026. Anyone calling this identifier needs to migrate.
What the data says. No volume over the last 7 days. Its weekly peak was 1.00%.
See WizardLM-2 8x22B on Model Season →Llama 3.3 Euryale 70B was removed from the catalog
Last listed on Oct 18, 2026. Anyone calling this identifier needs to migrate.
What the data says. No volume over the last 7 days. Its weekly peak was 0.14%.
See Llama 3.3 Euryale 70B on Model Season →Gemma 3 27B got 15% cheaper
From $0.17 to $0.15 per 1M tokens, blended. Input $0.080 → $0.080, output $0.45 → $0.35. New price since Oct 18.
What the data says. No volume over the last 7 days. Its weekly peak was 0.72%. Google has 3.8% of tokens and 5.4% of estimated spend this week. The lab's most used model is Gemini 3.8 Flash, with 1.07% at $1.65 per 1M, and this one costs 11 times less.
See Gemma 3 27B on Model Season →Other changes
- ContextLlama 4 Maverick changed its context window. From 1M to 512k tokens. See model →
- ContextMorph V3 Fast changed its context window. From 80k to 256k tokens.
In the news: summaries written by AI using only what the cited sources published, reviewed before going live. The numbers in the “The data” strip come from our data, traffic from the last 7 days, never from the summary. Catalog snapshot of Oct 19, 2026, compared with the previous ones since Oct 12, 2026. Traffic through Oct 18, 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.