AI model
DeepSeek's V4.1 Flash is a mixture-of-experts language model built for low-cost long-context serving, with a 1M-token window and a compact KV cache.
DeepSeek will return seven days of API spend as bonus credit after a 4h52m partial outage on October 15. The credit lands on a future balance and cannot answer a request that failed that morning, so failover to a second provider stays the app's own job.
Reality
GPT-6-Astra, running in OpenAI's Codex, solved 49 of 50 unpublished theoretical-physics problems on the TheoPhysEval benchmark. The problems are routine research calculations, and the authors say AI still lacks the "taste" for breakthroughs.
Reality
SpaceXAI lists Grok 4.7 at $2 and $6 per million tokens against $10 and $50 for GPT-6 Astra and Claude Fable 5.1, and the only per-task figures in the record put a Cursor run at $4.69 against about $9.
Perspective Coverage
8 publishersReality
The quarter-size global KV cache and eighth-size persistent storage are serving-cost claims rather than benchmark scores, and banking the second one requires a prefix-cache tier your stack has to already run.
Perspective Coverage
6 publishersReality
A single actor used a free Chinese pentest agent and four commercial AI models to breach South Korean financial firms in about two weeks. The free, resold toolkit let one person breach several firms at once, and researchers expect adversaries to keep using it to lift their tempo.
Reality
DeepSeek V4.1 Flash costs $2.10 off-peak for an agent run that costs $75 on Claude Opus 5, about 36x less, according to a dev.to analysis. Paying peak-hour rates halves that gap to about 18x, and running at max effort cuts it to about 15x.
Reality
DeepSeek is close to raising at least 80 billion yuan at a pre-money valuation of about 500 billion yuan, roughly $75 billion, Bloomberg reported. At about 75 times its $1 billion run rate, the round sets the floor a 2027 Shanghai STAR Market listing has to clear.
Perspective Coverage
17 publishersReality
DeepSeek released its open-source Harness agent runtime as a Windows and macOS desktop app on September 30, so users no longer need a command line to run it. The agent edits local files and runs commands, so what each user exposes to it decides what it can touch and which provider sees it.
Perspective Coverage
3 publishersReality
Hermes Agent v0.21.2 fills passwords from 1Password, Bitwarden or its own vault into web pages without the model ever seeing them. Every card fill waits for a person, so the agent cannot charge a stored card on its own, cron jobs included.
Reality
V4.1-Flash retires the V4 Pro line and carries two active-parameter counts, 763B total with 8B on input tokens and 16B on output, so one sizing number no longer covers both phases of a request. Baseten had it running on day zero.
Perspective Coverage
3 publishersReality
Flash's off-peak input price is under a quarter of what V4-Pro cost, and on DeepSeek's own table it beats the old Pro checkpoint on Terminal-Bench, but it scores 36.8 on Humanity's Last Exam and no V4.1-Pro has a date.
Reality
OpenRouter has dropped the stealth listing. The model is Pareto 26.9, from unbiased.ai, at $2.50 per million input tokens before a 10 October launch. That is double what 26.8 charged.
Reality
A dev.to post pulled every figure for 12 popular Claude Skills straight from the GitHub API on 19 September 2026, then set the result against what three skill catalogs advertise for the same repository.
Reality
The number came out of one harness on OpenRouter at temperature 0.9 with a 64,000-token output cap, and it holds for a bounded translation task while the same model sits mid-table on open-ended terminal work.
Reality
Nous Research spent about $19,300 on 1,393 subagents to take 34.4 percent out of Hermes's non-test Python in 19 hours. Test line count moved 0.06 percent, so the guardrail was the suite the team already had.
Reality
The vendor selling the cheapest model in the comparison reports a 0.7-point quality spread across four frontier models against run-to-run variation of 1.4 to 3.2 points. That leaves price per task, $0.43 against an implied $6.45 for GPT-6 Astra.
Reality
Z.ai's GLM 5.3 rejects every way of switching thinking off and defaults to max effort, where 33 of 33 checkable tasks came back right at $0.00468 per correct answer against $0.01173 from GLM 5.2.
Reality
DeepSeek's first CFO arrives with CITIC Securities engaged and a 2027 listing as the goal. The near-term cash is a 50 billion yuan private round, and the two published accounts of June's valuation differ by nearly seven times.
Reality
The V4.1-Flash change log puts Terminal Bench 2.1 at 90.6 against V4-Pro's 87.9 and DeepSWE at 74.2 against 62.7. DeepSeek says API prices came down with the release and points to a pricing page for the amounts.
Reality
Nvidia's Nemotron scored 30 of 42 at IMO 2026 writing proofs in ordinary prose, and the September 9 paper publishes the data, the code and the acceptance rule that decided which candidate proofs survived.
Reality