AISA-ArabicFC 2026
Models read an Arabic request and four tool schemas. They select a tool and supply its arguments, or decline to call a tool.
Saudi Legal FC ran a tuned Qwen2.5-7B-Instruct on the shared task. Organizers scored 21 teams on 1,125 blind requests.
- Made by
- Tuwaiq Academy
- Released
- May 2026
- Kind
- Function calling
- Level
- School
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Language
- Arabic
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
QMUL_AK tops the leaderboard with 0.91
Best score per entry from the benchmark's own leaderboard, read on November 20th, 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.
- 0.91 QMUL_AK
- 0.90 الشاهق
- 0.90 Ṣaqr (صقر)
- 0.90 Thaka
- 0.90 Yalla
- 0.88 Arabic Function Forge
- 0.88 PuDyDG
- 0.88 ArgTune Arabs
- 0.88 Sa74ll
- 0.87 Siraj
- 0.87 Durham-Jazan
- 0.87 وكلاء العربية
- 0.86 مَكين | Makeen
- 0.86 MOSKA-LLM
- 0.85 Saudi Legal FC
- 0.85 farah
- 0.83 GymMinds
- 0.82 ToolCallers
- 0.78 TIGER
- 0.77 واجه
- 0.69 Optimal
- 1 QMUL_AK 0.91
- 2 الشاهق 0.90
- 3 Ṣaqr (صقر) 0.90
- 4 Thaka 0.90
- 5 Yalla 0.90
- 6 Arabic Function Forge 0.88
- 7 PuDyDG 0.88
- 8 ArgTune Arabs 0.88
- 9 Sa74ll 0.88
- 10 Siraj 0.87
- 11 Durham-Jazan 0.87
- 12 وكلاء العربية 0.87
- 13 مَكين | Makeen 0.86
- 14 MOSKA-LLM 0.86
- 15 Saudi Legal FC 0.85
- 16 farah 0.85
- 17 GymMinds 0.83
- 18 ToolCallers 0.82
- 19 TIGER 0.78
- 20 واجه 0.77
- 21 Optimal 0.69
Source: Official AISA-ArabicFC 2026 leaderboard
| # | Entry | Score | Setting | Submitted by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | QMUL_AK | 0.91 | QMUL_AK | Official leaderboard | ||
| 2 | الشاهق | 0.90 | الشاهق | Official leaderboard | ||
| 3 | Ṣaqr (صقر) | 0.90 | Ṣaqr (صقر) | Official leaderboard | ||
| 4 | Thaka | 0.90 | Thaka | Official leaderboard | ||
| 5 | Yalla | 0.90 | Yalla | Official leaderboard | ||
| 6 | Arabic Function Forge | 0.88 | Arabic Function Forge | Official leaderboard | ||
| 7 | PuDyDG | 0.88 | PuDyDG | Official leaderboard | ||
| 8 | ArgTune Arabs | 0.88 | ArgTune Arabs | Official leaderboard | ||
| 9 | Sa74ll | 0.88 | Sa74ll | Official leaderboard | ||
| 10 | Siraj | 0.87 | Siraj | Official leaderboard | ||
| 11 | Durham-Jazan | 0.87 | Durham-Jazan | Official leaderboard | ||
| 12 | وكلاء العربية | 0.87 | وكلاء العربية | Official leaderboard | ||
| 13 | مَكين | Makeen | 0.86 | مَكين | Makeen | Official leaderboard | ||
| 14 | MOSKA-LLM | 0.86 | MOSKA-LLM | Official leaderboard | ||
| 15 | Saudi Legal FC | 0.85 | Saudi Legal FC | Official leaderboard | ||
| 16 | farah | 0.85 | farah | Official leaderboard | ||
| 17 | GymMinds | 0.83 | GymMinds | Official leaderboard | ||
| 18 | ToolCallers | 0.82 | ToolCallers | Official leaderboard | ||
| 19 | TIGER | 0.78 | TIGER | Official leaderboard | ||
| 20 | واجه | 0.77 | واجه | Official leaderboard | ||
| 21 | Optimal | 0.69 | Optimal | Official 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
- 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 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (21 entries).
- Pass Models scored 21 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 21 entries at the last read, none a model we track.
- 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.