How to host and improve the token speed of an LLM
So in this article, we will focus only on decreasing the TPS time.Experiment SetupHardware: Your options are wide in terms of choosing a GPU for this model to fit in. If you’re using a quantised NVFP4 version, you can choose RTX 3090 as your option.Model: Gemma 4 with 31 billion parameters, in its instruction tuned form. The first is the plain BF16 checkpoint, google/gemma-4-31B-it, which is the model exactly as released.
- ▪So in this article, we will focus only on decreasing the TPS time.Experiment SetupHardware: Your options are wide in terms of choosing a GPU for this model to fit in.
- ▪If you’re using a quantised NVFP4 version, you can choose RTX 3090 as your option.Model: Gemma 4 with 31 billion parameters, in its instruction tuned form.
- ▪The first is the plain BF16 checkpoint, google/gemma-4-31B-it, which is the model exactly as released.
Hacker News (AI / LLM) files mainly under ai. We currently carry 6,790 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Medium |
| Canonical URL | https://medium.com/@abhijithneilabraham/learning-inference-how-to-host-and-improve-the-token-speed-of-an-llm-cff5623ab505 |
| Publication time | Tue, 29 Sep 2026 08:03:21 +0000 |
| Retrieval time | 2026-09-29T08:14:55.890Z |
| Last seen | 2026-09-29T08:14:55.890Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | zu6HByBA6qT6 · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Llm InferenceArtificial IntelligenceInference EngineeringLLMMachine LearningLearning inference : How to host and improve the token speed of an LLMAbhijith Neil Abraham10 min read·Sep 19, 2026--2ListenShareInference engineering can be difficult in 2026, as there is no one step playbook yet fully solving best inference optimisations. Frameworks are still evolving to accomodate various model architectures, and this means this field and optimisation step requires understanding of GPU kernels, model configurations, architectures, ML theory and more.In this article, I will be using a 31B parameter Gemma 4 LLM to demonstrate how to improve the TPS (Tokens per second).TPS (Tokens per Second)TPS is about how fast the model generates tokens.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Medium.
Discussion
0 commentsMore from Medium
