ANALYTICS
A GPU-accelerated Bayesian GLM package buys its way into the standard R Bayesian toolchain
Alternatives · analytics tools
haze alternatives
The best haze alternatives in analytics tools, ranked by Sparkpulse's velocity_score.
Updated Aug 17, 2026
Looking for the best alternatives to haze? Sparkpulse tracks and ranks 12 alternatives in analytics tools by shipping velocity — how frequently each ships meaningful updates, verified from official changelogs. For reference, haze shipped 0 meaningful updates in the last 30 days and carries a velocity score of 2.5 out of 10 in 2026. The alternatives below are ranked the same way, so you're comparing real release momentum, not marketing claims.
About haze
Four dormant years end with a modernization pass and an off-by-one fix in the C++ core
haze does nearest-neighbour smoothing and k-d tree interpolation on brain surface meshes. It sat untouched from April 2022 until July 2026, when a single release modernized it for current R versions and corrected an off-by-one error in the C++ code. It is not on CRAN and never will be — the package exceeds 50MB against CRAN's 5MB ceiling, a constraint its own initial release notes acknowledge.
Velocity 2.5 · Last update 1h ago
Top 12 alternatives to haze
Ranked by recent ship velocity. Tap any card for the full editorial breakdown, or pivot to a head-to-head.
ANALYTICS
A single-cell data store commits to Zarr v3 and range-readable hosting across four language surfaces
ANALYTICS
A trend-extraction toolkit grows a full decomposition engine, seasonal components and all.
ANALYTICS
The standard QTL mapping package in R opened its genome scan to user-supplied likelihood models.
ANALYTICS
The R client for Our World in Data found its search had been reading a tenth of the catalog.
ANALYTICS
Text analysis in R keeps optimising its token internals — and builds a path out to torch
ANALYTICS
The ModernDive teaching package learns to render inside the browser that runs its own textbook
ANALYTICS
Continuous-time mediation effects get standardized centrality, six years into steady patch work
ANALYTICS
The explainable-ensemble-tree package now measures whether its own explanations are faithful.
ANALYTICS
The discrete-data FDR package is being pared into one piece of a larger multiple-testing suite.
ANALYTICS
A scientific-text analysis package moved from counting citations to classifying argument structure.
ANALYTICS
The teaching arm of an R reliability suite keeps pace with whatever its analysis siblings ship.
haze vs alternatives — shipping velocity at a glance
Velocity score (0–10) and meaningful releases shipped in the last 30 days, from official changelogs. Higher = shipping faster.
| Product | Velocity | Sparks · 30d | Focus areas | Latest release |
|---|---|---|---|---|
| haze (baseline) | 2.5 | 0 | neuroimagingmesh-processinginterpolation | — |
| glmbayes | 6.3 | 1 | bayesian-statisticsgeneralized-linear-modelsopencl | OpenCL split out to nmathopencl; insight and bayestestR integration |
| lstar | 5.0 | 0 | single-cell-genomicszarrwasm | lstar 0.2.0 |
| trendseries | 3.8 | 1 | time-serieseconometricsr-package | Decomposition becomes a first-class operation, five methods deep |
| qtl2 | 2.5 | 0 | qtl-mappingstatistical-geneticsbioinformatics | A genome scan that takes your own likelihood function |
| r-owidapi | 2.5 | 0 | open-dataour-world-in-datar-package | — |
| quanteda | 2.5 | 0 | text-analysisnatural-language-processingr-package | CRAN v4.0 |
| moderndive | 2.5 | 0 | statistics-educationwebrregression | — |
| cTMed | 2.5 | 0 | mediation-analysiscontinuous-time-modelsr-package | — |
| e2tree | 0.0 | 0 | explainable-aiensemble-methodsdecision-trees | A significance-tested measure of explanation fidelity |
| discretefdr | 0.0 | 0 | multiple-testingfalse-discovery-ratediscrete-statistics | — |
| contentanalysis | 0.0 | 0 | text-analysisbibliometricsscientific-writing | Sentence-level rhetorical move classification arrives |
| relialearnr | 0.0 | 0 | reliability-engineeringr-packageeducation | — |
The 12 best haze alternatives, in depth
1. glmbayes · velocity 6.3
A GPU-accelerated Bayesian GLM package buys its way into the standard R Bayesian toolchain.
Over the last 30 days glmbayes shipped 1 meaningful update vs haze's 0, most recently “OpenCL split out to nmathopencl; insight and bayestestR integration”. Its velocity score of 6.3/10 blends that with longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, glmbayes focuses on bayesian statistics, generalized linear models and opencl.
Over the last 30 days glmbayes has been shipping faster than haze — a point in its favour if release momentum matters to you.
2. lstar · velocity 5.0
A single-cell data store commits to Zarr v3 and range-readable hosting across four language surfaces.
Its velocity score of 5.0/10 reflects longer-term release cadence; its most recent meaningful update was “lstar 0.2.0”.
Where haze leans on neuroimaging, mesh processing and interpolation, lstar focuses on single cell genomics, zarr and wasm.
lstar and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
3. trendseries · velocity 3.8
A trend-extraction toolkit grows a full decomposition engine, seasonal components and all.
Over the last 30 days trendseries shipped 1 meaningful update vs haze's 0, most recently “Decomposition becomes a first-class operation, five methods deep”. Its velocity score of 3.8/10 blends that with longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, trendseries focuses on time series, econometrics and r package.
Over the last 30 days trendseries has been shipping faster than haze — a point in its favour if release momentum matters to you.
Full trendseries trajectory → · Compare haze vs trendseries →
4. qtl2 · velocity 2.5
The standard QTL mapping package in R opened its genome scan to user-supplied likelihood models.
Its velocity score of 2.5/10 reflects longer-term release cadence; its most recent meaningful update was “A genome scan that takes your own likelihood function”.
Where haze leans on neuroimaging, mesh processing and interpolation, qtl2 focuses on qtl mapping, statistical genetics and bioinformatics.
qtl2 and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
5. r-owidapi · velocity 2.5
The R client for Our World in Data found its search had been reading a tenth of the catalog.
Its velocity score of 2.5/10 reflects longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, r-owidapi focuses on open data, our world in data and r package.
r-owidapi and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
6. quanteda · velocity 2.5
Text analysis in R keeps optimising its token internals — and builds a path out to torch.
Its velocity score of 2.5/10 reflects longer-term release cadence; its most recent meaningful update was “CRAN v4.0”.
Where haze leans on neuroimaging, mesh processing and interpolation, quanteda focuses on text analysis, natural language processing and r package.
quanteda and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
7. moderndive · velocity 2.5
The ModernDive teaching package learns to render inside the browser that runs its own textbook.
Its velocity score of 2.5/10 reflects longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, moderndive focuses on statistics education, webr and regression.
moderndive and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
8. cTMed · velocity 2.5
Continuous-time mediation effects get standardized centrality, six years into steady patch work.
Its velocity score of 2.5/10 reflects longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, cTMed focuses on mediation analysis, continuous time models and r package.
cTMed and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
9. e2tree · velocity 0.0
The explainable-ensemble-tree package now measures whether its own explanations are faithful.
Its velocity score of 0.0/10 reflects longer-term release cadence; its most recent meaningful update was “A significance-tested measure of explanation fidelity”.
Where haze leans on neuroimaging, mesh processing and interpolation, e2tree focuses on explainable ai, ensemble methods and decision trees.
e2tree and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
10. discretefdr · velocity 0.0
The discrete-data FDR package is being pared into one piece of a larger multiple-testing suite.
Its velocity score of 0.0/10 reflects longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, discretefdr focuses on multiple testing, false discovery rate and discrete statistics.
discretefdr and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
Full discretefdr trajectory → · Compare haze vs discretefdr →
11. contentanalysis · velocity 0.0
A scientific-text analysis package moved from counting citations to classifying argument structure.
Its velocity score of 0.0/10 reflects longer-term release cadence; its most recent meaningful update was “Sentence-level rhetorical move classification arrives”.
Where haze leans on neuroimaging, mesh processing and interpolation, contentanalysis focuses on text analysis, bibliometrics and scientific writing.
contentanalysis and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
Full contentanalysis trajectory → · Compare haze vs contentanalysis →
12. relialearnr · velocity 0.0
The teaching arm of an R reliability suite keeps pace with whatever its analysis siblings ship.
Its velocity score of 0.0/10 reflects longer-term release cadence.
Where haze leans on neuroimaging, mesh processing and interpolation, relialearnr focuses on reliability engineering, r package and education.
relialearnr and haze have shipped at a similar pace over the last 30 days, so the decision comes down to fit and feature depth.
Full relialearnr trajectory → · Compare haze vs relialearnr →
Frequently asked questions
What are the best alternatives to haze?
The top haze alternatives we currently track in analytics tools are glmbayes, lstar, trendseries, qtl2, r-owidapi, ranked by recent ship velocity.
How is this list of haze alternatives ranked?
Alternatives are ranked by Sparkpulse's velocity_score — release cadence + 30-day spark count + sector-relative ship rate.
Can I compare haze directly with one of these alternatives?
Yes — every card has a "Compare with haze" link to a side-by-side /compare page.