Prices per 1M tokens.

API prices by model
ModelInputCache readOutputCapability (Epoch AI)
gemma-4-26B-A4B-it$0.14$0.05$0.40Capable · 141.6

Rates, billing conditions and sources

gemma-4-26B-A4B-it

Model ID: google/gemma-4-26B-A4B-it · Context window: 262K tokens

All rates, conditions and sources

Conditions of every rate below:

  • SiliconFlow international pricing page (www.siliconflow.com/pricing), LLM table: input, cached-input and output prices per 1 million tokens ("Prices shown are per 1 million tokens.").
  • Model "google/gemma-4-26B-A4B-it" (the SiliconFlow API model ID), the row "gemma-4-26B-A4B-it" whose Details link is the model page ./models/gemma-4-26b-a4b-it; that page was reviewed on 2026-10-16 to name this ID, which was in SiliconFlow's authenticated model list.
  • Currency: the page writes a bare "$" and names no currency; compute.info's owner confirmed on 2026-10-16 that these are US dollars.
  • Cache: SiliconFlow's pricing page, model pages and prompt-caching article (reviewed 2026-10-16 and 2026-10-17) state no cache duration, cache-write price or exact eligibility rule. A missing cache-read price is unknown, not zero.
  • List prices: the page states no region, endpoint, tier or account eligibility, and its FAQ offers volume discounts by arrangement.

Input: $0.14 per 1M tokens · Standard tier

  • Input tokens (the row's "Input (/M Tokens)" column).

Cache read: $0.05 per 1M tokens · Standard tier

  • Cached input tokens (the row's "Cached Input (/M Tokens)" column, labelled "Cache Read" on its model page): only eligible repeated input that hits SiliconFlow's prompt cache is billed at this rate.

Output: $0.40 per 1M tokens · Standard tier

  • Output tokens (the row's "Output (/M Tokens)" column).

Capability (Epoch AI): ECI 141.6 (90% range 139.1–144.0) · range spans bands · Epoch AI score for Gemma 4 26B A4B, model dated Apr 2, 2026

Price observed:

API availability: listed as available when checked (not a live service check)

Collection record

Primary source ↗ · provider documentation · Claims: siliconflow-gemma-4-26b-a4b-it: specifications · Retrieved · Checked · Reviewed · Freshness: unknown · Confidence: A

  • Official SiliconFlow model page for google/gemma-4-26B-A4B-it, spec block, raw SHA-256 44da84c555aef55c57a47859945baf676714d927ccdecda35e025d77aee96777 (the #104 identity review's copy, docs/provenance/siliconflow-identity-2026-10-16.json; the retrieval time is that copy's file time, an upper bound on the read): "Context length 262K".
  • Visible shorthand recorded as stated (#504); not converted to an exact token count. The pricing page's LLM table states the same value in its Context Length column; the SiliconFlow parser fails closed if it ever differs (#626).
  • Specifications only: no pricing, availability, capability or release-date assertion.

Primary source ↗ · provider documentation · Claims: siliconflow-gemma-4-26b-a4b-it: pricing · siliconflow-gemma-4-26b-a4b-it: availability · 52 more claims about 26 other catalog entries · Retrieved · Checked · Review due · Confidence: B

  • Raw SHA-256: 74219c83c4ffeeed010e2f82509ddd413f75f6376836e282985a75dea97c51ef; collector siliconflow-pricing-llm-table-1.
  • Hosted listing and pricing observation only: the reviewed LLM rows of the international pricing page; no checkpoint, developer, license, context-window, modality or release-date assertions. Like the other hosted providers, a listing is not proof of identical weights or behavior.
  • The page shows prices with a bare "$"; US dollars rests on the owner's confirmation of 2026-10-16 (#104), not on the page, hence confidence B. Its server-rendered LLM table is not a complete inventory (Load More), so an unlisted model is not evidence of anything.
About capability scores (Epoch AI)

Capability scores: Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource], epoch.ai/benchmarks. Accessed Nov 18, 2026. Licensed CC BY 4.0; values here are rounded to one decimal place and grouped into bands.

The Epoch Capabilities Index (ECI) combines many benchmarks on a scale where Claude 3.5 Sonnet is 130 and GPT-5 is 150. Bands, on the displayed one-decimal score: Frontier 160+, Strong 150–159.9, Capable 140–149.9, Entry below 140. Each score has a 90% range, listed with each model's rates and sources; a range that crosses a band boundary is flagged there. Models are not ranked: neighbouring scores often overlap.

A benchmark is not your workload. Epoch's scores mostly reflect runs at the highest reasoning effort, which use more output tokens than default settings. Scores are dated and can shift between Epoch releases. A score belongs to the developer's model; a host listing shows the same model's score, but hosts may serve different versions or quantizations.

How headline rates are chosen

Headline rates use the directory's shared rule: the rate a buyer meets by default, chosen by billing, tier and context dimensions and never by price. Context caps, the bracket starting at zero and short-context variants are labeled. Every price is a listed price, not a service guarantee; each offer's details say what its source states about availability. Batch, off-peak and other variants remain in the details; where no single rate is the default, the table shows a dash.

No currency conversion is performed. Missing cache prices are unknown, not zero. Equal units do not establish equivalent models or service conditions. See the methodology and source register.