Benchmark Almanac

Shop Assistant Honesty Bench (UA/RU)

Gemma 4 31B IT 100% leads since October 2026
Not in model cards

Models act as shop assistants in Ukrainian or Russian. The tests check language choice and whether quoted prices and stock match the catalog.

Kaggle publishes scores for 9 distinct model versions.

Made by
DimaTopka
Released
October 2026
Kind
Customer service
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Propensity
Behavior
Hallucination
Language
Ukrainian, Russian
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Occupations
Sales
Strongest model tested
Claude Sonnet 5.5 #9 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
9
Models scored on its own leaderboard

3 models share the top score of 100% among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemma 4 31B IT 100%
  2. 2 GPT-6 Luna 100%
  3. 3 Claude Sonnet 5.5 100%
  4. 4 gpt-oss-20b 98.1%
  5. 5 GLM-5 96.2%
  6. 6 Claude Haiku 4.5 84.6%
  7. 7 DeepSeek-R1 May 2025 78.8%
  8. 8 GPT-5.4 Mini 73.1%
  9. 9 Grok 4.20 non-reasoning 71.2%

Source: Official Shop Assistant Honesty Bench (UA/RU) leaderboard

#ModelScoreSettingReleasedRun onSource
1 Gemma 4 31B IT 100% Apr 2026 Oct 2026 Official leaderboard
2 GPT-6 Luna 100% Sep 2026 Oct 2026 Official leaderboard
3 Claude Sonnet 5.5 100% Sep 2026 Oct 2026 Official leaderboard
4 gpt-oss-20b 98.1% Aug 2025 Oct 2026 Official leaderboard
5 GLM-5 96.2% Feb 2026 Oct 2026 Official leaderboard
6 Claude Haiku 4.5 84.6% Oct 2025 Oct 2026 Official leaderboard
7 DeepSeek-R1 May 2025 78.8% May 2025 Oct 2026 Official leaderboard
8 GPT-5.4 Mini 73.1% Mar 2026 Oct 2026 Official leaderboard
9 Grok 4.20 71.2% non-reasoning Feb 2026 Oct 2026 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Own leaderboard Passed 85% in Oct 2026, by Gemma 4 31B IT

Source: Official Shop Assistant Honesty Bench (UA/RU) leaderboard

Gpt-oss-20b on the official leaderboard passed 85% in August 2025

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Gemma 4 31B IT gpt-oss-20b DeepSeek-R1 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Gemma 4 31B IT DeepSeek-R1

Source: Official Shop Assistant Honesty Bench (UA/RU) leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Tool use

All 265 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Neon 56.3%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (9 entries).
  • Pass Models scored 9 models we track.
  • Pass Newest score The newest is from September 2026, for Claude Sonnet 5.5.
  • Note Frontier models 1 of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Sonnet 5.5 (#9).
  • Note Headroom Saturated. The best is 100%, past 85% since August 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on kaggle.com: 9 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check