Benchmark Almanac

Snowflake Text-to-SQL HARD

Arctic Text2SQL R2 48.7% tops its leaderboard Gemini 3.1 Pro 47.9% leads the models we track
Not in model cards

Models generate SQL for difficult questions about unfamiliar enterprise schemas. Snowflake compares models by query accuracy.

The Arctic-Text2SQL-R2 release reports results for Gemini 3.1 Pro, Claude Opus 4.7 and Qwen3.6.

Made by
Snowflake
Released
May 2026
Kind
Data science and SQL
Level
Expert
Scored
Checked automatically
Answers
Code
Tasks
Kept private
Measures
Capability
Language
English
In and out
Text, Code → Code
Interaction
Single turn
Refresh
One fixed set
Run by
A host
In views
Law and finance
Strongest model tested
GPT-5.4 #28 on the capability index
Last model added
April 2026
Scores refreshed
November 21st, 2026
14
Entries on its own leaderboard, 12 models we track

Arctic Text2SQL R2 tops the leaderboard with 48.7% Gemini 3.1 Pro leads with 47.9% among models released by April 2026

Best score per entry from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Gemini 3.1 Pro, at 47.9%. This view shows only models we track, each at its best.

The newest model with a score here came out in April 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 Arctic Text2SQL R2 48.7%
  2. 2 Gemini 3.1 Pro 47.9%
  3. 3 Claude Opus 4.7 47.3%
  4. 4 Claude Sonnet 4.6 45.5%
  5. 5 Grok 4.20 45.5%
  6. 6 Arctic Text2SQL R1.5 45.1%
  7. 7 Gemini 3.1 Flash-Lite 44.5%
  8. 8 GPT-5.3 Codex 41.6%
  9. 9 Kimi K2.5 41%
  10. 10 Claude Haiku 4.5 40.8%
  11. 11 GPT-5.4 39%
  12. 12 GLM-5.1 38.2%
  13. 13 Grok 4.1 Fast 35.6%
  14. 14 Qwen 3.6 35B-A3B 33.7%

Source: Official Snowflake Text-to-SQL HARD leaderboard

#EntryScoreSettingDateSource
1 Arctic Text2SQL R2 48.7% Official leaderboard
2 Gemini 3.1 Pro 47.9% Feb 2026 Official leaderboard
3 Claude Opus 4.7 47.3% Apr 2026 Official leaderboard
4 Claude Sonnet 4.6 45.5% Feb 2026 Official leaderboard
5 Grok 4.20 45.5% Feb 2026 Official leaderboard
6 Arctic Text2SQL R1.5 45.1% Official leaderboard
7 Gemini 3.1 Flash-Lite 44.5% Mar 2026 Official leaderboard
8 GPT-5.3 Codex 41.6% Feb 2026 Official leaderboard
9 Kimi K2.5 41% Jan 2026 Official leaderboard
10 Claude Haiku 4.5 40.8% Oct 2025 Official leaderboard
11 GPT-5.4 39% Mar 2026 Official leaderboard
12 GLM-5.1 38.2% Apr 2026 Official leaderboard
13 Grok 4.1 Fast 35.6% Nov 2025 Official leaderboard
14 Qwen 3.6 35B-A3B 33.7% Apr 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3.1 Pro 47.9% Feb 2026 Official leaderboard
2 Claude Opus 4.7 47.3% Apr 2026 Official leaderboard
3 Claude Sonnet 4.6 45.5% Feb 2026 Official leaderboard
4 Grok 4.20 45.5% Feb 2026 Official leaderboard
5 Gemini 3.1 Flash-Lite 44.5% Mar 2026 Official leaderboard
6 GPT-5.3 Codex 41.6% Feb 2026 Official leaderboard
7 Kimi K2.5 41% Jan 2026 Official leaderboard
8 Claude Haiku 4.5 40.8% Oct 2025 Official leaderboard
9 GPT-5.4 39% Mar 2026 Official leaderboard
10 GLM-5.1 38.2% Apr 2026 Official leaderboard
11 Grok 4.1 Fast 35.6% Nov 2025 Official leaderboard
12 Qwen 3.6 35B-A3B 33.7% Apr 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 Not tested since May 2026

Source: Official Snowflake Text-to-SQL HARD leaderboard

Scores over time

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% 2026 Near the ceiling (85%) Gemini 3.1 Pro GPT-5.3 Codex Claude Haiku 4.5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro GPT-5.3 Codex

Source: Official Snowflake Text-to-SQL HARD 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 Coding

All 523 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for Claude Opus 4.7. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (14 entries).
  • Pass Models scored 12 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 7 months ago, in April 2026, for Claude Opus 4.7.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 (#28).
  • Pass Headroom The best is 47.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on snowflake.com: 12 rows for models we track and 2 other entries at the last read.
  • Pass Last read November 21st, 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