AgMoDB
Models
Agents
Evals
Explore
Tools
Industry
Menu
Gemma 3 12B Instruct | AgMoDB
All Models
Gemma 3 12B Instruct
March 12, 2025
Last synced Apr 7, 2026, 4:00 PM
Blended Price
Free/M
Input Price
Free/M
Output Price
Free/M
Speed
32 tok/s
TTFT
20.45s
Benchmark Scores
Intelligence Index
External Benchmarks
AA-Omniscience Hallucination Rate
Add to comparison
How Gemma 3 12B Instruct Compares
Axes
X Axis
Blended Price (USD)
Y Axis
AgMoBench Overall
Bubble Size
Context Window
Filters
Blended Price (USD)
$0.00 – $30.0
AgMoBench Overall
≥ 3.5
Providers
ai21-labs
alibaba
anthropic
aws
azure
baidu
cohere
deepseek
+15 more
Table view
Filters
Show quadrants
Top models
anthropic
openai
google
meta
mistral
nvidia
kimi
xai
azure
deepseek
aws
cohere
baidu
ai21-labs
zai
reka-ai
xiaomi
alibaba
minimax
ibm
Bubble size = Context Window
16384.00
2000000.00
Compare with other models
8.8
Coding Index
6.3
Math Index
18.3
MMLU Pro
0.6
/ 100
GPQA Diamond
0.3
/ 100
HLE
0.0
/ 100
LiveCodeBench
0.1
/ 100
SciCode
0.2
/ 100
MATH-500
0.9
/ 100
AIME
0.2
/ 30
AIME 2025
0.2
/ 30
IFBench
0.4
/ 100
LCR
0.1
/ 100
Terminal-Bench Hard
0.0
/ 100
τ²-Bench
0.1
/ 100
aa_omniscience
97.1
/ 100
AA-Omniscience Accuracy
Predicted
43.6
/ 100
Aider Polyglot
Predicted
33.2
/ 100
AlpacaEval 2.0 LC
Predicted
30.6
/ 100
AlpacaEval 2.0 Raw
Predicted
21.4
/ 100
ARC-AGI-1
Predicted
18.6
/ 100
ARC-AGI-1 Cost per Task
Predicted
0.1
ARC-AGI-2
Predicted
0.2
/ 100
ARC-AGI-2 Cost per Task
Predicted
0.1
BFCL (Berkeley Function Calling)
Predicted
35.8
BigCodeBench Complete
Predicted
45.8
/ 100
BigCodeBench Instruct
Predicted
36.6
/ 100
AA Intelligence Index (Matrix)
Predicted
61.7
AA Long Context Reasoning (Matrix)
Predicted
72.9
AIME 2024
Predicted
58.7
AIME 2025 (Matrix)
Predicted
78.8
Arena-Hard Auto
Predicted
38.1
BrowseComp
Predicted
70.6
BRUMO 2025
Predicted
95.0
CMIMC 2025
Predicted
82.5
CritPt
Predicted
0.4
GPQA Diamond (Matrix)
Predicted
63.9
GSM8K
Predicted
74.4
HLE (Matrix)
Predicted
44.6
HMMT Feb 2025
Predicted
82.9
HMMT Nov 2025
Predicted
91.1
HumanEval
Predicted
82.6
IFBench (Matrix)
Predicted
39.4
IFEval
Predicted
82.7
IMO 2025
Predicted
21.4
LiveCodeBench (Matrix)
Predicted
52.6
MATH-500 (Matrix)
Predicted
93.4
MathArena Apex 2025
Predicted
2.3
MMLU
Predicted
82.2
MMLU-Pro (Matrix)
Predicted
71.0
MMMU-Pro
Predicted
79.4
MRCR v2
Predicted
77.4
OSWorld
Predicted
38.8
SimpleQA
Predicted
26.2
SMT 2025
Predicted
85.6
SWE-bench Pro
Predicted
31.9
Tau-Bench Telecom (Matrix)
Predicted
97.2
Terminal-Bench 2.0
Predicted
36.2
Terminal-Bench 1.0
Predicted
15.4
USAMO 2025
Predicted
9.1
Video-MMU
Predicted
85.6
browsecomp
Predicted
69.8
BullshitBench
Predicted
22.5
/ 100
Aider Polyglot
Predicted
0.0
Apex Agents
Predicted
2.7
Arc Agi 2
Predicted
0.0
BALROG
Predicted
0.0
BIG-Bench Hard
Predicted
3.0
BoolQ
Predicted
0.8
CAD-Eval
Predicted
5.7
Chess Puzzles
Predicted
0.2
CyBench
Predicted
0.2
DeepResearchBench
Predicted
0.4
FictionLiveBench
Predicted
0.4
Gdpval
Predicted
0.3
GeoBench
Predicted
0.0
GSM8K (Epoch)
Predicted
5.0
GSO
Predicted
0.0
HellaSwag
Predicted
0.6
Hle
Predicted
0.1
Lech Mazur Writing
Predicted
7.3
METR Time Horizons
Predicted
1.6
OTIS Mock AIME 2024–2025
Predicted
0.1
PIQA
Predicted
0.9
Posttrainbench
Predicted
0.0
SimpleQA Verified (Epoch)
Predicted
0.3
The Agent Company
Predicted
1.4
TriviaQA
Predicted
5.4
VPCT
Predicted
0.4
WinoGrande
Predicted
0.8
FrontierMath
Predicted
19.7
/ 100
GAIA Level 1
Predicted
32.9
GAIA Level 2
Predicted
10.4
GAIA Level 3
Predicted
0.2
GAIA
Predicted
20.7
/ 100
LegalBench
Predicted
70.1
/ 100
LiveBench Coding
Predicted
73.4
/ 100
LiveBench Data Analysis
Predicted
80.5
/ 100
LiveBench Language
Predicted
84.6
/ 100
LiveBench Math
Predicted
61.0
/ 100
LiveBench Overall
Predicted
81.3
/ 100
LiveBench Reasoning
Predicted
80.4
/ 100
LongBench v2 Easy
Predicted
41.6
LongBench v2 Hard
Predicted
37.0
LongBench v2
Predicted
33.7
/ 100
MathVista
Predicted
48.5
/ 100
MLE-bench
Predicted
32.2
/ 100
MMMU
Predicted
67.4
/ 100
MMTU Table Understanding
Predicted
45.5
/ 100
MT-Bench
Predicted
7.8
/ 10
OCRBench v2
Predicted
75.2
/ 100
Open LLM Average
Predicted
28.1
/ 100
Open LLM: BBH
Predicted
55.7
/ 100
Open LLM: GPQA
Predicted
31.4
/ 100
Open LLM: IFEval
Predicted
61.8
/ 100
Open LLM: MATH Level 5
Predicted
20.0
/ 100
Open LLM: MMLU-PRO
Predicted
39.5
/ 100
Open LLM: MUSR
Predicted
43.6
/ 100
RE-Bench
Predicted
47.8
SimpleBench
Predicted
33.1
/ 100
simpleqa
Predicted
29.2
SWE-bench Lite
Predicted
18.1
/ 100
SWE-bench Verified
Predicted
46.8
/ 100
tau-bench Retail
Predicted
70.9
/ 100
Vectara Factual Consistency
Predicted
95.4
/ 100
Vectara Hallucination Rate
Predicted
4.6
/ 100
WeirdML
Predicted
16.7
/ 100
WildBench
Predicted
42.7
Chatbot Arena ELO
chatbot_arena
4203.0
NoLiMa (NIAH)
nolima
87.4
/ 100
WebArena
webarena
26.1
/ 100