DeepSeek

In the repliesThe author clarifies that AgenC used peak pricing for DeepSeek’s estimate; an off-peak calculation would be about $0.43. Alternative_Web7202 questions the lack of evaluation criteria and calls the post promotion for the harness. Other commenters dispute mechanical details in the animations.

One shared-harness comparison with unequal stopping conditions. The post links the tool; the author’s financial affiliation is unverified. Token totals include repeated and cached context across calls, not just generated output. No simulation or bill was independently reproduced.

Original posted October 24 at 18:16:37 UTC. The post and author pricing clarification were read October 26, alongside the reply questioning evaluation criteria. Cost figures remain estimates.

In the repliesIn reply 2, the author says comparison runs used the same prompt, fresh sessions and rolled-back source code. On October 22, reply 17 says they submitted a refund request and again saw overload errors during short translation requests. That is a refund request, not confirmation of payment back.

One user’s incident across different clients and routes. Reasoning settings and full task logs are missing; the post does not establish a provider-wide outage or prove that the model changed after its preview.

Original October 21 at 02:38 UTC; numbered author replies 2 and 17 read October 23. Other replies give mostly unmatched negative impressions, not a controlled reproduction.

In the repliesNo human replies were visible in the opened launch thread. The linked repository describes a beta with three levels and user-supplied model access; its default 200-call limit caps requests, not the dollar cost of a round.

This is a creator’s project promotion and early comparison, not an independent ranking. The current README documents checks with Jev and DeepSeek; other model routes are not all verified. No gameplay or latency test was reproduced for this summary.

Original October 21 at 12:07 UTC; original and linked README read October 23. The current repository permits a consented server proxy for Jev, alongside direct browser-to-provider routes.

In the repliesdom96 counters that the model performed around DeepSeek V4 Pro on their own adversarial programming-language benchmark. LaurensBER asks for task completion time alongside low token prices, while explicitly saying they have not tried the new model.

The comments mix early-preview experiences, a personal benchmark and broader evaluation preferences. Exact hosted revisions, prompts and matched final-release tests are missing; benchmark manipulation is not established.

Story posted October 19 at 16:39:09 UTC. Comments 50157109, 50157373, 50159325 and 50155706 checked through HN’s official API; no independent reproduction.

In the repliesSubject-Mine4885 says they also use GLM for coding but have not tried dividing roles. 9kSs asks which coding tool or agent framework is being used. BusinessData5321 highlights file conflicts and asks whether the audit catches model mistakes or mainly integration problems.

The post does not specify the coding tool or agent framework or establish measured speed, accuracy or savings. The author describes some review models as free through WorkBuddy; that does not establish general API pricing.

Original posted October 18 at 14:21:07 UTC. Human comments phi3rqu, phi4wxg and phjq7sa inspected October 19; automated moderation excluded.

In the repliesjiabubu asks which reasoning intensity was used, a material missing configuration detail. johnlanni suggests comparing the combined time and cost of review, repair and re-review against an all-Sol run; that commenter also discloses maintaining a routing project.

One developer’s task and assessment, not an independently reproduced model ranking. The posted token-change percentages are internally unclear and are not used here.

Original October 18 and selected human replies 5 and 8 opened October 21; promotional project link omitted. No test executed by TechReadly.

In the repliespoofph reported that Flash Next could apply a fix in five minutes and then spend an hour testing it, and described using prompts to limit extra tests and verbosity. BeaveItToLeever used a modified DeepSeek Harness with medium and occasional extra-high effort, reporting higher Qwen throughput after tuning on a different setup. These were separate experiences, not a verified fix for the author's workflow.

Mac Studio M1 Ultra, 128GB. The author did not provide matched quantizations, runtime, reasoning effort, task logs or an exact DeepSeek revision. Token throughput and end-to-end task duration measure different things.

Original October 14 post and human replies pgrlu6u and pgsautl checked October 15. Reports were not independently reproduced.

In the repliesPackageDecent3526 recommended Sonnet 5.5 for architectural intent and edge cases. Cyvster suggested Qwen, reporting that a local 9B model had found issues in DeepSeek-written code. These were individual recommendations rather than a comparison on the author’s repository.

Exact Kimi and Qwen versions, prompts, client, hardware and API plans were not provided. The reported fixes and code quality were not independently tested.

Original October 5 post and human replies pe22fk4 and pe6vsd0; self-reported model names and experience. No general model ranking follows.

In the repliesThe comment listing was empty when checked; no reply-based confirmation or challenge was available.

These are one author’s small-task results, not a complete model ranking or an independently reproduced benchmark. Newly generated tasks do not establish freedom from contamination.

Original post reviewed October 6. Strata is linked in the post; the author’s relationship to the project is unconfirmed. Raw execution traces were not inspected.

In the repliesCommenter ricedumpling described a separate coding workflow: use GPT Plus to develop a plan, save business logic and interface details as Markdown in the project, then have Claude Code connected to the official DeepSeek API read that plan and execute. When zzbd questioned the cache cost of switching models, ricedumpling clarified that GPT served as a consultant while execution and input caching stayed on DeepSeek; no measured savings were supplied.

The author did not identify a DeepSeek version. Evening use is a personal routine, not a verified discount schedule.

Original post and named replies on LINUX DO; spending and workflow results are self-reported, not independently tested.