Loading company intelligence…
Loading the latest published snapshot and its supporting evidence.
LIVE RANKINGS & DATA
Loading the latest published snapshot and its supporting evidence.
Exact Qwen3-8B (Prompt) configuration in the reviewed BFCL v4 leaderboard.
A current roll-up of published ranks, normalized facts, primary evidence, and entity relationships. No missing value is inferred.
| Metric | Value | As of | Evidence | Confidence |
|---|---|---|---|---|
| model_tool_useBFCL evaluation total cost | $63.95reported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL format-sensitivity max delta | 16.5 percentage_pointsreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL format-sensitivity standard deviation | 5.09 percentage_pointsreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL irrelevance-detection accuracy | 82.27 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL live accuracy | 80.09 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL mean latency | 54.17 secondsmeasured | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL memory accuracy | 13.12 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL multi-turn accuracy | 33.38 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL non-live AST accuracy | 88.56 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL overall accuracy | 40.43 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL p95 latency | 194.15 secondsmeasured | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL relevance-detection accuracy | 75 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
| model_tool_useBFCL web-search accuracy | 13.5 percentreported | Apr 13, 2026 | Berkeley Function Calling Leaderboard ↗Grade A | 100% |
Every currently published position, joined to its snapshot, confidence, and methodology version.
| Rank | Leaderboard | Score | Movement | Confidence | As of | Method | Open explanation |
|---|---|---|---|---|---|---|---|
| #44 | BFCL Tool-Use Configurationsbfcl-tool-use-configurations | 40.43 | — | 100% | Apr 13, 2026 | vbfcl-v4-2026-04-13-official-v1-bfcl-tool-use-configurations-v1 | Explain → |
Ranks exact model-plus-inference configurations by official UC Berkeley BFCL overall accuracy.
Some explanation inputs are unavailable Missing: publicFrozenEvidence, comparablePriorRanking.
| 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 |
13 published observations for Qwen3-8B (Prompt). Open a row for the source, evidence location, extraction method, and quality state.
| Metric | Value | Kind | As of | Confidence | Evidence |
|---|---|---|---|---|---|
| BFCL web-search accuracybfcl-web-search-accuracy | 13.5percent | reported | Apr 13, 2026 | 100% | View source
|
| BFCL overall accuracybfcl-overall-accuracy | 40.43percent | reported | Apr 13, 2026 | 100% | View source
|
| — |
| 0% |
| 0 |
| 1 |
| available |
| 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% | 40.43 | 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 | Qwen | 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 | Qwen3-8B (Prompt) | Unavailable | — | 0 | unavailableno_public_frozen_evidence |
|---|
| upstreamRank | 44 | Unavailable | 44 | 0 | unavailableno_public_frozen_evidence |
|---|
no_prior_published_snapshot
No comparable component drivers were published.
{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}{"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"}| BFCL live accuracybfcl-live-accuracy | 80.09percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL format-sensitivity standard deviationbfcl-format-sensitivity-standard-deviation | 5.09percentage_points | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL format-sensitivity max deltabfcl-format-sensitivity-max-delta | 16.5percentage_points | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL multi-turn accuracybfcl-multi-turn-accuracy | 33.38percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL mean latencybfcl-latency-mean-seconds | 54.17seconds | measured | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL irrelevance-detection accuracybfcl-irrelevance-detection-accuracy | 82.27percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL evaluation total costbfcl-evaluation-total-cost-usd | 63.95USD · USD | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL non-live AST accuracybfcl-non-live-ast-accuracy | 88.56percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL relevance-detection accuracybfcl-relevance-detection-accuracy | 75percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL memory accuracybfcl-memory-accuracy | 13.12percent | reported | Apr 13, 2026 | 100% | View source
|
|---|
| BFCL p95 latencybfcl-latency-p95-seconds | 194.15seconds | measured | Apr 13, 2026 | 100% | View source
|
|---|