Gemma 1.1 7B It
Mar 2024
Gemma 1.1 7B ItCompare inon
Benchmarks of Gemma 1.1 7B It14
BenchmarkByScoreRank
No benchmark matches
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGemma 3 12BunstatedOpenwins 74%-11%$0.05 / $0.10Memory (4-bit): 4.3 GB → 6.1 GB
BetterGPT-5 Minihighwins 73%+882%$0.053CheaperGPT OSS 120BunstatedOpenwins 68%-14%$0.03 / $0.15Memory (4-bit): 4.3 GB → 70 GBCheaperGemma 3 4BunstatedOpenwins 67%-39%$0.04 / $0.05CheaperGemma 2 9B ItOpenwins 65%-46%$0.03 / $0.06Memory (4-bit): 4.3 GB → 4.6 GBModelScoreCostPer taskWhat changes
Gemma 3 12BunstatedOpenwins 74%-11%$0.05 / $0.10Memory (4-bit): 4.3 GB → 6.1 GBGPT OSS 120BunstatedOpenwins 68%-14%$0.03 / $0.15Memory (4-bit): 4.3 GB → 70 GBGemma 3 4BunstatedOpenwins 67%-39%$0.04 / $0.05Gemma 2 9B ItOpenwins 65%-46%$0.03 / $0.06Memory (4-bit): 4.3 GB → 4.6 GBGPT OSS 20BunstatedOpenwins 63%-49%$0.018 / $0.09Memory (4-bit): 4.3 GB → 13 GBModelScoreCostPer taskWhat changes
GPT-5 Minihighwins 73%+882%$0.053GPT-6 Lunamaxwins 82%+186%$0.068Function calling only with none · Chat Completions+3GPT-5.6 Lunaxhighwins 82%+543%$0.085GPT-5.6 Terraxhighwins 83%+6,329%$0.63Prompts with >272K input tokensGPT-6.1 Solmaxwins 85%+5,614%$0.72Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
Gemma 3 12BunstatedOpenwins 74%-11%$0.05 / $0.10Memory (4-bit): 4.3 GB → 6.1 GBGemma 3n E4B InstructunstatedOpenwins 70%——
NVIDIA Nemotron 3.5 Lightning 30B A3B NVFP4Openwins 68%——Memory (4-bit): 4.3 GB → 11 GBGemma 3 4BunstatedOpenwins 67%-39%$0.04 / $0.05
Granite 4.1 8BunstatedOpenwins 66%——Memory (4-bit): 4.3 GB → 4.4 GBModelScoreCostPer taskWhat changes
Gemini 4 Argonhighwins 89%—*$1.99Synchronous output stops at 65,536+2Gemini 3.8 Flashhighwins 87%—*$1.24Sampling parameters are ignored+1Gemini 3.1 Pro Previewunstatedwins 87%—$2 / $12Cannot use both thinking controlsGemini 3.6 Flashhighwins 86%—*$1.60Gemini 3 Flash Previewunstatedwins 85%—$0.50 / $3Gemini 2.5 Prounstatedwins 84%—$1.25 / $10Gemini 3.5 Flash Liteunstatedwins 84%—$0.30 / $2.50Gemma 4 31BunstatedOpenwins 82%—$0.10 / $0.34Memory (4-bit): 4.3 GB → 19 GBAll Google models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · compressed✓ 16 GB · 8-bit✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceGemma termscommercial use with conditions
Benchmarks
all categoriesChat & Knowledge
Arena Text · overall1,183#185
Arena Text · Chinese1,060.6#174
Arena Text · English1,124#181
Arena Text · French1,065.5#146
Arena Text · German1,053.7#149
Arena Text · instruction following1,056.7#182
Arena Text · Japanese970.7#132
Arena Text · Korean987.9#139
Arena Text · non-English1,051.5#180
Arena Text · Russian1,045.8#179
Arena Text · Spanish1,049.1#154
Arena Text · overall · full cohort1,093.9#181
Writing & Language
Arena Text · creative writing1,150#184
Arena Text · creative writing · full cohort1,059.3#181
FAQ
How much does Gemma 1.1 7B It cost?›
$0.07 per 1M input tokens and $0.07 per 1M output tokens (cheapest host: DeepInfra).
Is Gemma 1.1 7B It free to use?›
There is no free API; the cheapest paid price is $0.07 per 1M input tokens. The weights are open, so you can run it yourself.
How good is Gemma 1.1 7B It, and what is it best at?›
#185 of 196 in Chat & Knowledge, #184 of 195 in Writing & Language.
Is Gemma 1.1 7B It better than Mixtral 8x7B Instruct v0.1?›
On Arena Text · creative writing, Gemma 1.1 7B It scores 1,150 vs 1,159 — about the same.
How can I use Gemma 1.1 7B It?›
Through the API of DeepInfra.
What is the context window of Gemma 1.1 7B It?›
8K tokens.
When was Gemma 1.1 7B It released? Is there a newer version?›
Released 2024-03-26. It is the current version in its line.
Is Gemma 1.1 7B It open source?›
Yes, the weights are open (gemma licence).
Can I run Gemma 1.1 7B It locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can Gemma 1.1 7B It read images?›
No, it accepts text only.
What is Gemma 1.1 7B It?›
Gemma 1.1 7B It is an open-weights language model from Google, released Mar 2024.
What are the best alternatives to Gemma 1.1 7B It?›
Gemma 3 12B (cheaper), GPT-5 Mini (better).