# DeepSeek ships V4.1-Flash-1017, MIT-licensed, 1M context

> DeepSeek released V4.1-Flash-1017, an MIT-licensed update with a 1M-token context and higher benchmark scores.

*The open-weight update lifts reasoning, coding and agentic scores over the original V4.1 Flash, though Opus 5 still leads on most benchmarks.*

By Bez · FeaturedDaily
Canonical: https://featureddaily.com/news/deepseek-ships-v4-1-flash-1017-mit-licensed-1m-context

**What happened:** DeepSeek released V4.1-Flash-1017 on Hugging Face. The repository and weights are under the MIT licence.

**The details:** The [model card](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) calls it a minor update to V4.1-Flash. The context window is 1M tokens. DeepSeek recommends a max_tokens of 256K or more, with temperature 1.0 and top_p 1.0.

**The numbers:** Flash-1017 against the original Flash:

HLE without tools: 42.6 vs 36.8. Terminal-Bench 4.0: 42.3 vs 31.2. DeepSWE v1.1: 77.0 vs 74.2.

ProgramBench: 29.1 vs 20.3. Apex Bench v1.1: 72.1 vs 64.6. BabyVision: 92.5 vs 89.6.

**In their words:** The model card describes "substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks" (DeepSeek, model card). It also says the model cuts task execution time and output length, which lowers cost.

**The rivals:** The card also compares the model with K3, GLM-5.3, GPT-5.6 Sol and Opus 5. Opus 5 scores higher than Flash-1017 on most of the listed benchmarks.

**The catch:** There is no Jinja chat template. DeepSeek instead provides a Python reference implementation (encoding.py) and a Rust toolkit, deepseek-recipe, for production use.

**Unconfirmed:** The model card gives no release date. The 17 October date comes from a news roundup on aiweekly.co. That roundup also lists 552B total parameters and new Search and Fetch APIs at $0.75 per 1,000 successful requests. The model page shows 763B, and it lists no API pricing.

**What's next:** DeepSeek has not said what follows. Developers can pull the weights from Hugging Face now.

## Key takeaways

- MIT-licensed weights, 1M-token context window
- Beats original Flash on every listed benchmark
- Opus 5 still scores higher on most

## Sources

- [DeepSeek-V4.1-Flash-1017 model card](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) — Hugging Face (DeepSeek), 2026-10-17