explainx.ai350K+ learn from us
TrendingAI News TodayExplainx for Business
Pricing
explainx.ai

Upskill in AI — 16 free pathways, live workshops & bootcamps, and 50+ courses from practitioners. Plus the skills, tools, and MCP servers to practice on.

follow us

follow on google

Add explainx.ai as a preferred source

corporate training

support@explainx.ai

get started

Find your pathTake Free Evaluation

community

Join the community

learn

mind: share how you thinkpathways — start freeworkshopsbootcampscoursescompare Explainxcertificationsmock testsexplainx universitycorporate traininglearn skills & mcp

discover

skillsmcp serversexplainx mcpestacktoolsmdx readeragentsllmsdesignsdictionarypeopleagi trackerfelony benchranks

company

aboutvisionmissionteaminstructorsteach on explainxpartnershipscommunityhackathonscareers

content

daily AI newsstate of AI — live resultsblogreleasespromptsgeneratorsresource libraryfor LLMsexplainx.ai kids

solutions

all solutionsdeveloper upskillingmarketing upskillingproduct manager upskillingleadership upskilling

newsletter · weekly

Get AI news, tools, and insights in your inbox.

supportcontactprivacytermsdata rightshow we create contentsubmission guidelines

© 2026 AISOLO Technologies Pvt Ltd

explainx.ai

On this page

  • Quick answer
  • What decision models do
  • Why Gemma 4 backbones now
  • Calibration and soft labels
  • 128k context use cases
  • Model pick guide (from Baseten table)
  • How to test without hype
  • Competitive landscape
  • Sovereignty and open weights
  • Pricing snapshot
  • API migration from mammoth-1
  • Multimodal decisions
  • Open weights operations
  • Artificial Analysis vs your labels
  • Pareto frontier claim
  • Agent orchestration fit
  • Failure modes to log in production
  • Gemma 4 supply chain notes
  • groundhog-1 positioning
  • Building a decision-model evaluation harness
  • Security and prompt injection on decision inputs
  • Quick glossary for buyers
  • Related reading
← Back to blog

explainx / blog

Baseten bison-1: Gemma 4 Decision Model Tops AA Index

Baseten, Decision Models, Open Weights, Gemma

Part of Decision Models

Base Labs shipped bison-1 and groundhog-1 on Nov 20, 2026: open Gemma 4 decision models with 128k context, multimodal input, and #1 Artificial Analysis Decision score.

Nov 20, 2026·8 min read·Yash Thakker
add explainx.ai
go deep
Baseten bison-1: Gemma 4 Decision Model Tops AA Index

Base Labs shipped bison-1 and groundhog-1 on November 20, 2026, open decision models built on Google Gemma 4. In the launch post, Baseten says bison-1 ranks #1 of 19 models on Artificial Analysis Decision Index v2.1 with the best Brier calibration on that leaderboard.

Quick answer

table · 6 cols
ModelParamsBackboneAPI contextPrice / 1M tokensIndex claim
bison-131BGemma 4 31B128k$0.1076, #1 per Baseten
groundhog-126B MoE (~4B active)Gemma 4 26B-A4B128k$0.035~mammoth-1 class expected
mammoth-1 (prior)27BQwen16k$0.1071 on same index
elephant-1 (prior)9BQwen16k$0.035Lower tier

Artificial Analysis Decision Index ranking with bison-1 at number oneArtificial Analysis Decision Index ranking with bison-1 at number one Image: Baseten

Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

What decision models do

Decision models sit between cheap classifiers and frontier chat models. You feed state (text and optional images) plus a question with discrete options. The model returns probabilities, not a prose essay.

That pattern fits:

  • Support ticket routing when policy paragraphs live in the context.
  • Risk scoring when you need calibrated confidence for automation gates.
  • Agent orchestration when each step is a small choice, as OpenAI Performance Factory and Microsoft-Decision-1.1 at Ignite describe for routing at scale.

Baseten’s earlier Liquid AI d1-ultra coverage explained the economics of sub-cent decision APIs. bison-1 is pricier per token but claims a higher index score and Gemma lineage.

Why Gemma 4 backbones now

Last month Baseten released mammoth-1 and elephant-1 on Qwen, then said teams asked for Western open bases for sovereignty and procurement reasons.

bison-1 and groundhog-1 answer that without resetting the API shape: change the model name on the same Baseten decision endpoint.

The post also claims the family now sits on the Pareto frontier of accuracy vs time from elephant-1 through bison-1 on Artificial Analysis charts—vendor graphics, not independent replication here.

Calibration and soft labels

Brier score calibration chart for bison-1 at 0.14 on the Decision IndexBrier score calibration chart for bison-1 at 0.14 on the Decision Index Image: Baseten

Baseten highlights Brier score 0.14 for bison-1 versus 0.15 for d1.2-ultra on the Decision Index. Brier measures how well stated confidence matches empirical accuracy.

Training changes include more soft labels (example: teach 83% A / 17% B instead of hard 100% labels). That matters when your workflow sends only high-confidence decisions to straight-through processing.

Compare with Sarvam Prism open confidence models if you self-host instead of using Baseten APIs.

128k context use cases

Earlier Base Labs decision APIs capped at 16k tokens. 128k lets you include full policy PDFs, email threads, or multi-image packets without pre-summarization.

Baseten’s example: an invoice looks payable until an earlier cancellation email in the same thread changes the decision. Long context reduces wrong approvals when the decisive sentence is far from the question.

Watch latency: larger contexts still cost time even if per-token prices look small.

Model pick guide (from Baseten table)

table · 2 cols
ChooseWhen
bison-1Highest accuracy needs, long context, budget for $0.10/1M
groundhog-1Long context with elephant-class pricing
mammoth-1Qwen backbone acceptable, knowledge-heavy decisions
elephant-1Default low-latency tier
joey-1Extreme cost sensitivity at 2B scale

Weights for bison-1 and groundhog-1 are on Hugging Face per the post, with MIT license listed on the library page.

How to test without hype

  1. Baseline your current router (rules, embeddings, or chat model).
  2. Mirror traffic on Baseten with bison-1 on a sample of real tickets.
  3. Compare ECE/Brier on your labels, not only Artificial Analysis aggregates.
  4. Load test 128k inputs—verify p95 latency fits SLA.
  5. Check modality needs; both models accept text and images.

If you run Qwen harness adapters for coding agents, keep decision models on a separate endpoint so you do not conflate codegen and routing metrics.

Competitive landscape

Artificial Analysis is the named index in the launch. Community trackers such as Jev Decision Index on Metatext rank many open decision fine-tunes separately; scores differ by benchmark version.

Recent explainx.ai decision coverage includes:

  • Liquid d1.2-ultra audio tasks
  • Anthropic overtakes OpenAI on Ramp AI Index for spend, not decision calibration
  • Microsoft-Decision-1.1 at Ignite for cheap routing inside Microsoft stack

Sovereignty and open weights

Open Gemma 4 weights let teams audit post-training data cards and host inference inside their VPC. Baseten still offers managed APIs for teams without GPU ops.

Pair with OVH Olala European open MoE discussions when RFPs demand EU or US model lineage.

Pricing snapshot

table · 2 cols
API model$/1M tokens (Baseten post)
bison-10.10
groundhog-10.035
mammoth-10.10
elephant-10.035

Decision workloads with many tiny calls should model monthly call volume, not only token rates.

API migration from mammoth-1

Baseten stresses API compatibility: swap the model string on existing decision endpoints. That lowers migration cost for teams already on mammoth-1 or elephant-1.

Migration checklist:

  1. Shadow groundhog-1 on latency-sensitive paths while bison-1 handles hard cases.
  2. Log probabilities and compare to prior Qwen-backed outputs on held-out tickets.
  3. Revisit thresholds; calibration shifts can change auto-approval rates even when accuracy rises.
  4. Update data residency paperwork to cite Gemma 4 lineage and Hugging Face weight locations.

Multimodal decisions

Image plus text input matters when decisions depend on scanned forms, damage photos, or UI screenshots. bison-1 accepts both modalities per the library page.

Run ablation tests with text-only inputs to see how much vision actually contributes; some teams pay multimodal latency without gains when OCR text is already extracted.

Open weights operations

MIT-licensed weights on Hugging Face enable on-prem or VPC deploys. You still need GPU ops, batching, and monitoring Baseten otherwise sells as managed service.

For European deployments, compare against OVH Olala and ResolveAI open SRE weights if your problem is incident routing rather than generic classification.

Artificial Analysis vs your labels

The 76 index score is vendor-reported alongside Artificial Analysis. Community indexes like Jev Decision Index rank hundreds of fine-tunes with different scores.

Before marketing “#1 decision model,” run your labels through bison-1 and d1.2-ultra on the same API latency tier. Index leaders change whenever benchmark versions update.

Pareto frontier claim

Decision Index score versus time per task for bison-1, mammoth-1, and elephant-1Decision Index score versus time per task for bison-1, mammoth-1, and elephant-1 Image: Baseten

Baseten’s blog includes a accuracy vs time chart placing bison-1, mammoth-1, and elephant-1 on a Pareto line. Charts illustrate positioning; they are not independent audits.

Use them to shortlist candidates, not to skip load tests.

Agent orchestration fit

Decision models excel when agents need dozens of micro-choices per task. Pair bison-1 with:

  • OpenAI Performance Factory style routing fixes.
  • Replit missions swarm economics when cost caps matter.
  • Ploy gateway speed claims only after you measure end-to-end with your decision layer included.

Failure modes to log in production

Decision APIs fail quietly when option sets are misspecified or images exceed tokenizer limits. Log:

  • Entropy of output distributions (flat probabilities often mean ambiguous prompts).
  • Refusal or null options when schemas allow abstain.
  • Latency vs context length coupling on 128k inputs.

Add human escalation when confidence sits in the middle band; calibrated models still err on tail events.

Gemma 4 supply chain notes

Choosing Gemma 4 backbones ties you to Google’s license and safety card updates for the base model. Monitor Gemma security advisories separately from Baseten API changelogs.

If procurement bans Qwen lineage but allows Gemma, document that choice in vendor security reviews alongside Cursor Spacex Chinese open-weight model policy threads customers now ask about.

groundhog-1 positioning

groundhog-1 targets elephant-class pricing with mammoth-class accuracy expectations on Gemma. Baseten says it should match mammoth-1 on the official AA index when scores publish.

Use groundhog-1 when p95 latency dominates SLAs and you still need 128k context—invoice thread example from the vendor post.

Building a decision-model evaluation harness

Treat Baseten’s index rank as hypothesis, not proof:

  1. Export 1,000 historical decisions with human labels.
  2. Format each as state + question + options identical to production schema.
  3. Run bison-1, mammoth-1, and a cheap baseline (rules or embeddings).
  4. Report accuracy, Brier, ECE, p95 latency, and dollars per 1M tokens on the same chart.
  5. Roll to 5% canary traffic before full cutover.

Without step four, you risk paying 31B inference where 9B elephant-1 already suffices.

Security and prompt injection on decision inputs

Long-context decision models ingest untrusted email bodies or user-uploaded scans. Attackers can hide instructions far from the question.

Mitigations mirror agent security elsewhere: strip HTML, hash known templates, require structured sender metadata, and cap image inputs from anonymous users. For deeper agent guardrails see Skyflow AgentCore interceptor patterns, adapted to your stack.

Re-run evaluations quarterly; Gemma and Qwen backbones will keep shipping, and Baseten’s menagerie will add new names that move the Pareto line again.

Quick glossary for buyers

table · 2 cols
TermMeaning here
Decision modelReturns probabilities over options, not chat prose
Brier scoreCalibration metric; lower is better on the AA index
Decision IndexArtificial Analysis composite for this model class
Soft labelsTraining targets like 83/17 splits instead of hard 100%

Related reading

  • Liquid d1-Ultra decision model economics
  • Liquid d1.2-ultra on decision audio tasks
  • Sarvam Prism calibrated decision weights
  • OpenAI Performance Factory routing fixes
  • Microsoft Ignite decision and agent analytics
  • Hugging Face harness adapters for coding
Spotted something out of date? Let us know.
Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

  • ✓ Free
  • ✓ One email a day, no spam
  • ✓ Unsubscribe anytime
Yash Thakker

Written by

Yash Thakker

Yash is an AI expert with over 300K learners. Join his workshops →

View Yash Thakker in People in AI →

Related posts

Nov 6, 2026

Nace AI Drex: Open-Weight Decision Models, and the Licence Catch

Nace AI has put its Drex decision models on open weights. Drex Max v1 is a dense 27B model that scores typed questions instead of writing prose. Here is what is confirmed, what is self-reported, and why "open source" is the wrong label.

Oct 23, 2026

Fastino GLiDE v1.2 Claims #1 on the Decision Index at 71.5: What Is Verified

Fastino Labs released GLiDE v1.2 on October 22, 2026, claiming first place on the Hugging Face Decision Index with 71.5, lower calibration error and new image input. The weights are on Hugging Face, but they are large and carry a non-standard licence. Here is what we could check.

Oct 22, 2026

Base Labs Mammoth-1 and Elephant-1: Open Decision Models Claim #1 (Claimed vs Verified)

Base Labs, part of Baseten, released mammoth-1, a 27B decision model, and its 9B sibling elephant-1 with weights and a technical report. The model card says mammoth-1 tops the Jev Decision Index and a replica of the Artificial Analysis index. Here is what is confirmed.