Gemma 3 4B IT
GoogleKey gemma-3-4b-it
No public Tokscale usage yet
No public Tokscale submission includes Gemma 3 4B IT yet, so this page has no active users, tokens or agent shares. The benchmark results, list prices and specs below come from public sources. See models with usage
Benchmarks
The best result for Gemma 3 4B IT in each benchmark Tokscale tracks. Each benchmark has its own unit, run conditions and source, so compare a score only with other models on the same benchmark. Rank is among the models Tokscale matched in that benchmark.
- Epoch Capabilities IndexOverall · Epoch AI115.9
Epoch Capabilities Index: 115.9
- Model
- Gemma 3 4B IT
- Listed as
- Gemma 3 4B
- Conditions
- Not reported
- Interval
- 97.4–123.6
- Unit
- ECI
- Model released
- Mar 12, 2025
- Retrieved
- Oct 21, 2026
Source · Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks (CC BY 4.0)
Rank #158 of 160 matched modelsNo conditions reportedModel released Mar 12, 2025Source: epoch.ai (opens in a new tab) - GPQA DiamondReasoning · Epoch AI23.2%
GPQA Diamond: 23.2%
- Model
- Gemma 3 4B IT
- Listed as
- gemma-3-4b-it
- Conditions
- Not reported
- Unit
- % correct
- Run date
- Aug 28, 2026
- Retrieved
- Oct 21, 2026
Source · Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks (CC BY 4.0)
Rank #149 of 151 matched modelsNo conditions reportedRun date Aug 28, 2026Source: epoch.ai (opens in a new tab)Near its ceiling: the top score is 96.4%, and Epoch AI estimates the reachable maximum at 97%. Differences between top models are small.
Pricing
List prices from models.dev in US dollars per 1M tokens, retrieved Oct 21, 2026. Amazon Bedrock is the reference price for the estimates below. It is the first provider listed; models.dev lists no price from the lab's own API.
Price at real agent usageEstimate at list prices
What 1M tokens cost at the reference list price when they split into input, output and cache the way Tokscale users' agents used them. It is not what anyone paid.
- At the community mix
- $0.04
- Est. per 1M tokens
96% cache reads - 3:1 input/output blend
- $0.05
- Per 1M tokens at list prices
No cache
Cache reads were 96% of all tokens in public Tokscale submissions in the last 30 days. The 3:1 blend, (3 × input + output) ÷ 4, is the usual comparison and ignores cache.
Assumptions
- Amazon Bedrock list prices per 1M tokens: input $0.04, output $0.08.
- The community mix is the tokens of every identified model in the last 30 days.
- This is an estimate at list prices, not what anyone paid.
- Reasoning tokens are priced as output.
- Long-context tiers, batch pricing and discounts are not applied.
- Token mixes come from public Tokscale submissions. Hidden accounts are excluded.
- No cache-read price is listed, so cache reads are priced as input.
- No cache-write price is listed, so cache writes are priced as input.
List prices by provider
- Amazon BedrockReference
google.gemma-3-4b-it$0.04 input per 1M tokens$0.08 output per 1M tokensCache read not listedCache write not listed131K tokens of context - OpenRouter
google/gemma-3-4b-it$0.04 input per 1M tokens$0.05 output per 1M tokens$0.02 cache read per 1M tokensCache write not listed131K tokens of context - Deep Infra
google/gemma-3-4b-it$0.05 input per 1M tokens$0.10 output per 1M tokensCache read not listedCache write not listed131K tokens of context
Specs
From models.dev (MIT), retrieved Oct 21, 2026, as the lab or the first listed provider reports them. A dash means models.dev does not list the value.
- Context window
- 131K tokens
- Max output
- 118K tokens
- Input
- Text, Image
- Output
- Text
- Knowledge cutoff
- Aug 2024
- Release date
- Mar 12, 2025
- Open weights
- Yes
- Reasoning
- No
- Tool calling
- No
Sources and attribution
Each score keeps its own source, unit and run conditions. Tokscale does not combine benchmarks into one score, and a missing score means the model was not evaluated, not that it scored zero.