Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

razorback16 openjev diffusiongemma (NVFP4, vLLM)

razorback16Decision modelOpen weights

razorback16 openjev diffusiongemma (NVFP4, vLLM) by razorback16 is a decision model: it answers typed questions with a probability for each allowed answer instead of generating text. It scores 40.0 on the Jev Decision Index (rank 82).

Pricing

No hosted price is listed. The weights are open: the cost is your own compute.

Specifications

Provider
razorback16
Parameters
26B
Released
Sep 2026
Open weights
Yes
Question types
Choice (one of N options), Score (ordered rubric), Yes / no
Mechanism
Diffusion LM
Backbone
google/diffusiongemma-26B-A4B-it
Jev-compatible request
No
Runs on CPU
No
License
—

Benchmarks

Jev Decision IndexECE 0.232
40.0
Jev Decision Indexedition 0.1 (earlier edition)
35.6
DecisionBenchECE 0.228
0.684
Jev Decision Index — VisionECE 0.174
47.7

Jev Decision Index: public 37.8, ECE 0.232, Brier 0.600. Lower ECE means its probabilities can be taken at face value.

Speed & latency

Median latency per call (Decision Index)
38 ms

Compare razorback16 openjev diffusiongemma (NVFP4, vLLM) with…

razorback16 openjev diffusiongemma (NVFP4, vLLM) vs amos-1-large—/1M inrazorback16 openjev diffusiongemma (NVFP4, vLLM) vs amos-1-medium—/1M inrazorback16 openjev diffusiongemma (NVFP4, vLLM) vs Celeris-1 Decision$0.04/1M inrazorback16 openjev diffusiongemma (NVFP4, vLLM) vs ScaleDown Classify-1$0.04/1M inrazorback16 openjev diffusiongemma (NVFP4, vLLM) vs Cloudflare clef—/1M inrazorback16 openjev diffusiongemma (NVFP4, vLLM) vs Cloudflare clef-flash—/1M in

FAQ

How much does razorback16 openjev diffusiongemma (NVFP4, vLLM) cost?

razorback16 openjev diffusiongemma (NVFP4, vLLM) has no hosted price listed; its weights are open, so the cost is the compute you run it on.

What questions can razorback16 openjev diffusiongemma (NVFP4, vLLM) answer?

razorback16 openjev diffusiongemma (NVFP4, vLLM) answers choice (one of n options), score (ordered rubric), yes / no questions, returning a probability for each allowed answer.

Is razorback16 openjev diffusiongemma (NVFP4, vLLM) well calibrated?

On the Jev Decision Index its expected calibration error (ECE) is 0.232; below about 0.05, a stated 80% confidence is right roughly 80% of the time.

Is razorback16 openjev diffusiongemma (NVFP4, vLLM) open weight, and can it run on a CPU?

Yes, razorback16 openjev diffusiongemma (NVFP4, vLLM) has open weights. It needs a GPU to run at useful speed.

Data as of 2026-10-14. Source: Metatext decision-model catalogue, 2026-10-14 (Jev Decision Index, DecisionBench). Pricing is per 1M tokens (USD); confirm with the provider before production use.