SYNCING VERIFIED SOURCES

Loading ranking table…

Loading the latest published snapshot and its supporting evidence.

BFCL Tool-Use Configurations | Rank.ai

RANK.AI DATA / Latest

BFCL Tool-Use Configurations

Exact model-plus-inference configurations ranked by the official UC Berkeley BFCL v4 overall accuracy.

Coverage: 109 entities
Confidence: 100%
Freshness: current
Export: CSV / JSON

PUBLIC VIEW / NO LOGINTrack this exact boardFilters and comparisons stay in the URL. Bookmark it, share it, or follow verified snapshot changes.

Follow changes

Watching publication Apr 13, 2026

LEADERBOARD

Current rankings

Metric: BFCL overall accuracy

Find in tableEntity typeMovementConfidenceEvidence NRank rangeColumns

109of 109 rows

BFCL Tool-Use Configurations current rankings
Rank	Entity	Signal	BFCL overall accuracy	Confidence	N	Explain
01	Claude-Opus-4-5-20251101 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	77.47%	100%	11	Why this rank
02	Claude-Sonnet-4-5-20250929 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	73.24%	100%	11	Why this rank
03	Gemini-3-Pro-Preview (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	72.51%	100%	13	Why this rank
04	GLM-4.6 (FC thinking)model	Official UC Berkeley BFCL v4 leaderboard.	72.38%	100%	11	Why this rank
05	Grok-4-1-fast-reasoning (FC)model	Official UC Berkeley BFCL v4 leaderboard.	69.57%	100%	11	Why this rank
06	Claude-Haiku-4-5-20251001 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	68.70%	100%	11	Why this rank
07	Gemini-3-Pro-Preview (FC)model	Official UC Berkeley BFCL v4 leaderboard.	68.14%	100%	11	Why this rank
08	o3-2025-04-16 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	63.05%	100%	13	Why this rank
09	Grok-4-0709 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	62.97%	100%	13	Why this rank
10	Grok-4-0709 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	61.38%	100%	11	Why this rank
11	Moonshotai-Kimi-K2-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	59.06%	100%	11	Why this rank
12	Grok-4-1-fast-non-reasoning (FC)model	Official UC Berkeley BFCL v4 leaderboard.	58.29%	100%	11	Why this rank
13	Command A Reasoning (FC)model	Official UC Berkeley BFCL v4 leaderboard.	57.06%	100%	11	Why this rank
14	DeepSeek-V3.2-Exp (Prompt + Thinking)model	Official UC Berkeley BFCL v4 leaderboard.	56.73%	100%	13	Why this rank
15	Gemini-2.5-Flash (FC)model	Official UC Berkeley BFCL v4 leaderboard.	56.24%	100%	11	Why this rank
16	GPT-5.2-2025-12-11 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	55.87%	100%	11	Why this rank
17	GPT-5-mini-2025-08-07 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	55.46%	100%	11	Why this rank
18	xLAM-2-32b-fc-r (FC)model	Official UC Berkeley BFCL v4 leaderboard.	54.66%	100%	11	Why this rank
19	DeepSeek-V3.2-Exp (FC)model	Official UC Berkeley BFCL v4 leaderboard.	54.12%	100%	11	Why this rank
20	GPT-4.1-2025-04-14 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	53.96%	100%	11	Why this rank
21	o4-mini-2025-04-16 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	53.24%	100%	11	Why this rank
22	xLAM-2-70b-fc-r (FC)model	Official UC Berkeley BFCL v4 leaderboard.	53.07%	100%	11	Why this rank
23	Qwen3-235B-A22B-Instruct-2507 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	52.15%	100%	13	Why this rank
24	GPT-5-nano-2025-08-07 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	51.45%	100%	11	Why this rank
25	Nanbeige4-3B-Thinking-2511 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	51.40%	100%	11	Why this rank
26	Gemini-2.5-Flash (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	50.90%	100%	13	Why this rank
27	GPT-4.1-mini-2025-04-14 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	50.45%	100%	11	Why this rank
28	o4-mini-2025-04-16 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	50.26%	100%	13	Why this rank
29	Qwen3-32B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	48.71%	100%	11	Why this rank
30	o3-2025-04-16 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	48.56%	100%	11	Why this rank
31	Qwen3-235B-A22B-Instruct-2507 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	47.99%	100%	11	Why this rank
32	Nanbeige3.5-Pro-Thinking (FC)model	Official UC Berkeley BFCL v4 leaderboard.	47.68%	100%	11	Why this rank
33	Qwen3-32B (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	46.78%	100%	13	Why this rank
34	xLAM-2-8b-fc-r (FC)model	Official UC Berkeley BFCL v4 leaderboard.	46.68%	100%	11	Why this rank
35	Command A (FC)model	Official UC Berkeley BFCL v4 leaderboard.	46.49%	100%	11	Why this rank
36	BitAgent-Bounty-8Bmodel	Official UC Berkeley BFCL v4 leaderboard.	46.23%	100%	11	Why this rank
37	Arch-Agent-32Bmodel	Official UC Berkeley BFCL v4 leaderboard.	45.37%	100%	11	Why this rank
38	GPT-5.2-2025-12-11 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	45.27%	100%	13	Why this rank
39	Qwen3-8B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	42.57%	100%	11	Why this rank
40	ToolACE-2-8B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	42.44%	100%	13	Why this rank
41	Qwen3-30B-A3B-Instruct-2507 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	41.39%	100%	11	Why this rank
42	xLAM-2-3b-fc-r (FC)model	Official UC Berkeley BFCL v4 leaderboard.	41.22%	100%	11	Why this rank
43	Qwen3-14B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	41.03%	100%	11	Why this rank
44	Qwen3-8B (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	40.43%	100%	13	Why this rank
45	GPT-4.1-2025-04-14 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	39.38%	100%	13	Why this rank
46	mistral-large-2411 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	38.37%	100%	11	Why this rank
47	Qwen3-14B (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	37.77%	100%	13	Why this rank
48	Mistral-Medium-2505model	Official UC Berkeley BFCL v4 leaderboard.	37.69%	100%	13	Why this rank
49	Mistral-Medium-2505 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	37.56%	100%	11	Why this rank
50	Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	37.29%	100%	11	Why this rank
51	Mistral-small-2506 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	37.15%	100%	11	Why this rank
52	Gemini-2.5-Flash-Lite (FC)model	Official UC Berkeley BFCL v4 leaderboard.	36.87%	100%	11	Why this rank
53	Qwen3-30B-A3B-Instruct-2507 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	36.70%	100%	13	Why this rank
54	Qwen3-4B-Instruct-2507 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	35.68%	100%	11	Why this rank
55	Qwen3-4B-Instruct-2507 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	35.52%	100%	13	Why this rank
56	Arch-Agent-3Bmodel	Official UC Berkeley BFCL v4 leaderboard.	35.36%	100%	11	Why this rank
57	Claude-Opus-4-5-20251101 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	33.47%	100%	13	Why this rank
58	GPT-4.1-nano-2025-04-14 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	33.05%	100%	11	Why this rank
59	Mistral-Small-2506 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	32.38%	100%	13	Why this rank
60	Arch-Agent-1.5Bmodel	Official UC Berkeley BFCL v4 leaderboard.	32.14%	100%	11	Why this rank
61	Command R7B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	32.07%	100%	11	Why this rank
62	Llama-3.3-70B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	31.90%	100%	11	Why this rank
63	mistral-large-2411 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	31.84%	100%	13	Why this rank
64	Hammer2.1-7b (FC)model	Official UC Berkeley BFCL v4 leaderboard.	31.67%	100%	11	Why this rank
65	xLAM-2-1b-fc-r (FC)model	Official UC Berkeley BFCL v4 leaderboard.	30.44%	100%	11	Why this rank
66	Gemma-3-12b-it (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	30.43%	100%	13	Why this rank
67	GPT-4.1-mini-2025-04-14 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	29.73%	100%	13	Why this rank
68	Hammer2.1-3b (FC)model	Official UC Berkeley BFCL v4 leaderboard.	29.71%	100%	11	Why this rank
69	Gemma-3-27b-it (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	29.47%	100%	13	Why this rank
70	Phi-4 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	28.79%	100%	13	Viewing why
71	Qwen3-1.7B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	28.41%	100%	11	Why this rank
72	Llama-4-Scout-17B-16E-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	28.13%	100%	11	Why this rank
73	Gemini-2.5-Flash-Lite (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	28.03%	100%	13	Why this rank
74	CoALM-70Bmodel	Official UC Berkeley BFCL v4 leaderboard.	27.99%	100%	13	Why this rank
75	Hammer2.1-1.5b (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.88%	100%	11	Why this rank
76	palmyra-x-004 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.87%	100%	11	Why this rank
77	GPT-5-mini-2025-08-07 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	27.83%	100%	13	Why this rank
78	Open-Mistral-Nemo-2407 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.63%	100%	11	Why this rank
79	GPT-5-nano-2025-08-07 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	27.55%	100%	13	Why this rank
80	Amazon-Nova-2-Lite-v1:0 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.10%	100%	11	Why this rank
81	Granite-3.1-8B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.10%	100%	11	Why this rank
82	Falcon3-10B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	27.01%	100%	11	Why this rank
83	Granite-3.2-8B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	26.87%	100%	11	Why this rank
84	CoALM-8Bmodel	Official UC Berkeley BFCL v4 leaderboard.	26.81%	100%	13	Why this rank
85	Llama-3.1-8B-Instruct (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	25.83%	100%	13	Why this rank
86	MiniCPM3-4B-FC (FC)model	Official UC Berkeley BFCL v4 leaderboard.	25.55%	100%	11	Why this rank
87	Claude-Haiku-4-5-20251001 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	25.26%	100%	13	Why this rank
88	Amazon-Nova-Pro-v1:0 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	24.97%	100%	11	Why this rank
89	Claude-Sonnet-4-5-20250929 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	24.90%	100%	13	Why this rank
90	GPT-4.1-nano-2025-04-14 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	24.88%	100%	13	Why this rank
91	Falcon3-7B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	24.03%	100%	11	Why this rank
92	Qwen3-0.6B (FC)model	Official UC Berkeley BFCL v4 leaderboard.	23.93%	100%	11	Why this rank
93	Granite-20b-FunctionCalling (FC)model	Official UC Berkeley BFCL v4 leaderboard.	23.23%	100%	11	Why this rank
94	Qwen3-0.6B (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	22.38%	100%	13	Why this rank
95	Amazon-Nova-Micro-v1:0 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	22.29%	100%	11	Why this rank
96	RZN-T (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	22.25%	100%	13	Why this rank
97	MiniCPM3-4B (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	22.08%	100%	13	Why this rank
98	Llama-3.2-3B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	21.95%	100%	11	Why this rank
99	Bielik-11B-v2.3-Instruct (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	21.90%	100%	13	Why this rank
100	Hammer2.1-0.5b (FC)model	Official UC Berkeley BFCL v4 leaderboard.	21.22%	100%	11	Why this rank
101	Gemma-3-4b-it (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	19.62%	100%	13	Why this rank
102	Open-Mistral-Nemo-2407 (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	19.31%	100%	13	Why this rank
103	Granite-4.0-350m (FC)model	Official UC Berkeley BFCL v4 leaderboard.	18.98%	100%	11	Why this rank
104	Falcon3-3B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	16.25%	100%	11	Why this rank
105	Ministral-8B-Instruct-2410 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	11.10%	100%	13	Why this rank
106	Falcon3-1B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	11.08%	100%	11	Why this rank
107	Llama-3.2-1B-Instruct (FC)model	Official UC Berkeley BFCL v4 leaderboard.	10.82%	100%	11	Why this rank
108	Llama-3.1-Nemotron-Ultra-253B-v1 (FC)model	Official UC Berkeley BFCL v4 leaderboard.	10.00%	100%	11	Why this rank
109	Gemma-3-1b-it (Prompt)model	Official UC Berkeley BFCL v4 leaderboard.	7.17%	100%	13	Why this rank

HISTORICAL SERIES

BFCL Tool-Use Configurations leader

Historical data is not available for this snapshot yet.

Some explanation inputs are unavailable Missing: publicFrozenEvidence, comparablePriorRanking.

Published rank: #70
Score: 28.79
Overall confidence: 100%
Coverage: 62%
Sample size: 13
As of: Apr 13, 2026, 11:59 PM UTC

SCORE CONSTRUCTION

Component ledger

13/21 evidenced

Components contributing to Phi-4 (Prompt)'s rank
Component	Value	Weight	Contribution	Evidence	Status
benchmarkVersion	v4	Unavailable	—	0	unavailableno_public_frozen_evidence
bfcl evaluation total cost usd	—	0%	0	1	available
bfcl format sensitivity max delta	—	0%	0	1	available
bfcl format sensitivity standard deviation

bfcl irrelevance detection accuracy	—	0%	0	1	available

bfcl latency mean seconds	—	0%	0	1	available

bfcl latency p95 seconds	—	0%	0	1	available

bfcl live accuracy	—	0%	0	1	available

bfcl memory accuracy	—	0%	0	1	available

bfcl multi turn accuracy	—	0%	0	1	available

bfcl non live ast accuracy	—	0%	0	1	available

bfcl overall accuracy	—	100%	28.79	1	available

bfcl relevance detection accuracy	—	0%	0	1	available

bfcl web search accuracy	—	0%	0	1	available

interactionMode	Prompt	Unavailable	—	0	unavailableno_public_frozen_evidence

metrics	Structured	Unavailable	—	0	unavailableno_public_frozen_evidence

organization	Microsoft	Unavailable	—	0	unavailableno_public_frozen_evidence

rankingDirection	descending	Unavailable	—	0	unavailableno_public_frozen_evidence

reviewedAsOf	2026-04-13	Unavailable	—	0	unavailableno_public_frozen_evidence

sourceConfiguration	Phi-4 (Prompt)	Unavailable	—	0	unavailableno_public_frozen_evidence

upstreamRank	70	Unavailable	70	0	unavailableno_public_frozen_evidence

PRIMARY EVIDENCE

Source trail

13 public records

BFCL evaluation total costbfcl evaluation total cost usd · observed Apr 13, 2026, 11:59 PM UTC8.72 USD100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL format-sensitivity max deltabfcl format sensitivity max delta · observed Apr 13, 2026, 11:59 PM UTC81.5 percentage_points100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL format-sensitivity standard deviationbfcl format sensitivity standard deviation · observed Apr 13, 2026, 11:59 PM UTC23.34 percentage_points100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL irrelevance-detection accuracybfcl irrelevance detection accuracy · observed Apr 13, 2026, 11:59 PM UTC87.55 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL mean latencybfcl latency mean seconds · observed Apr 13, 2026, 11:59 PM UTC9.49 seconds100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL p95 latencybfcl latency p95 seconds · observed Apr 13, 2026, 11:59 PM UTC23.02 seconds100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL live accuracybfcl live accuracy · observed Apr 13, 2026, 11:59 PM UTC60.7 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL memory accuracybfcl memory accuracy · observed Apr 13, 2026, 11:59 PM UTC24.73 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL multi-turn accuracybfcl multi turn accuracy · observed Apr 13, 2026, 11:59 PM UTC3.88 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL non-live AST accuracybfcl non live ast accuracy · observed Apr 13, 2026, 11:59 PM UTC69.56 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL overall accuracybfcl overall accuracy · observed Apr 13, 2026, 11:59 PM UTC28.79 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL relevance-detection accuracybfcl relevance detection accuracy · observed Apr 13, 2026, 11:59 PM UTC50 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗
BFCL web-search accuracybfcl web search accuracy · observed Apr 13, 2026, 11:59 PM UTC4.5 percent100% confidence
Source
Berkeley Function Calling Leaderboard
Grade
A
Freshness
fresh
Locator
{"csv":"https://gorilla.cs.berkeley.edu/data_overall.csv","repository":"https://github.com/ShishirPatil/gorilla/tree/main/berkeley-function-call-leaderboard","leaderboard":"https://gorilla.cs.berkeley.edu/leaderboard.html","rowIdentity":"Model","rankingColumn":"Overall Acc"}
Open primary evidence ↗