Skip to content
Model Pareto

Gemma 4 31B

Google DeepMindOpen weightsblog.google ↗
Compare with nearest rivalsCompare with…
Input
$0.14/M tok
Output
$0.40/M tok
Blended
$0.205/M tok · 3:1
Speed
35tok/s
Context
256Ktokens

Open weights; no paid Google API tier, price is the Artificial Analysis third-party host price.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.

Hardware
1× B200 180GB
FP8 weights
Throughput
~5,049 tok/s
many requests batched
Price
$0.16–1.41 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, ~82 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (30.7B params) + 25% KV-cache headroom ≈ 38 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~5,049 output tok/s aggregate at batch 144 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~34,629 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

SWE-bench VerifiedEstimated
~#9 of 14 (estimated)~75.5%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. #8Muse Glimmer76.0%
  4. ~#9Gemma 4 31BEstimated75.5%
  5. #9Xing4.0-29B-A4B75.0%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#14 of 19 (estimated)~54.8%
  1. #1Claude Opus 5.589.9%
  2. #2Claude Sonnet 5.581.3%
  3. #13Hy357.9%
  4. ~#14Gemma 4 31BEstimated54.8%
  5. #14Inkling54.3%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#27 of 31 (estimated)~51.1%
  1. #1DeepSeek V4.1 Flash74.2%
  2. #2Grok 4.772.6%
  3. #26Nex-N2.5-Pro55.8%
  4. ~#27Gemma 4 31BEstimated51.1%
  5. #27Qwen3.8 Max (0902)51.0%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~last of 15 (estimated)below measured range (<41.1)
  1. #1Claude Opus 5.566.0
  2. #2Claude Fable 5.162.2
  3. #14GPT-6 Luna41.1
  4. ~#15Gemma 4 31BEstimated<41.1
No published score — estimated from related benchmarks

Writing & chat

LMArena Text, Non-English (Elo)Few votesVerified
#24 of 531437 Elo
  1. #1Claude Opus 5.51501
  2. #2Claude Fable 5.1Few votes1496
  3. #23GPT-5.4 mini1439
  4. #24Gemma 4 31BFew votes1437
  5. #25GPT-6 Luna1436
lmarena.ai · 2026-09-24
LMArena Text (Elo)Verified
#23 of 531451 Elo
  1. #1Claude Opus 5.51509
  2. #2Claude Fable 5.11501
  3. #22Grok 4.61453
  4. #23Gemma 4 31B1451
  5. #24GPT-5.4 mini1448
lmarena.ai · 2026-09-24

Vision

MMMUEstimated
~#2 of 3 (estimated)~71.3%
  1. #1Claude Haiku 4.573.2%
  2. ~#2Gemma 4 31BEstimated71.3%
  3. #2Llama 4 Maverick69.4%
No published score — estimated from related benchmarks
CharXiv ReasoningEstimated
~#7 of 9 (estimated)~78.0%
  1. #1Gemini 3.8 Flash86.2%
  2. #2Kimi K384.8%
  3. #6Inkling78.1%
  4. ~#7Gemma 4 31BEstimated78.0%
  5. #7Claude Sonnet 577.0%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#5 of 8 (estimated)~83.9%
  1. #1MiniMax-M391.6%
  2. #2Kimi K391.1%
  3. #4Gemini 3.1 Pro85.3%
  4. ~#5Gemma 4 31BEstimated83.9%
  5. #5Claude Haiku 4.579.6%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#28 of 43 (estimated)~12.2%
  1. #1GPT-6 Astra31.0%
  2. #2Muse Spark 1.326.6%
  3. #26Qwen3.7 Plus12.2%
  4. ~#28Gemma 4 31BEstimated12.2%
  5. #28Muse Glimmer10.0%
No published score — estimated from related benchmarks
ChartographyEstimated
~#15 of 23 (estimated)~18.5%
  1. #1GPT-6 Astra71.0%
  2. #2Claude Opus 5.566.3%
  3. #14Grok 4.620.5%
  4. ~#15Gemma 4 31BEstimated18.5%
  5. #15Muse Glimmer17.2%
No published score — estimated from related benchmarks

Hard reasoning

AA Intelligence IndexEstimated
~#37 of 75 (estimated)~24.6
  1. #1Claude Opus 5.557.6
  2. #2Claude Sonnet 5.556.0
  3. #36Inkling25.0
  4. ~#37Gemma 4 31BEstimated24.6
  5. #37Ling-3.0-flash-VL24.6
No published score — estimated from related benchmarks

Agents

OSWorld-VerifiedEstimated
~#10 of 12 (estimated)~66.6%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #9Nex-N2.5-Mini71.2%
  4. ~#10Gemma 4 31BEstimated66.6%
  5. #10Muse Glimmer65.9%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #4GPT-5.547.5%
  4. ~#5Gemma 4 31BEstimated<47.5%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#15 of 18 (estimated)~60.4%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #14Inkling77.1%
  4. ~#15Gemma 4 31BEstimated60.4%
  5. #15Mistral Medium 3.548.6%
No published score — estimated from related benchmarks