Kaelum BenchBack to calibration results

Public Calibration

IFRS Reporting

Kaelum Bench · FMB-KNOW-ACC-IFRS-02

This view covers the concrete 60-task suite only. Its provisional 600-point score uses deterministic verification only. It exposes sanitized score, execution, token, and cost aggregates, with no prompts, responses, expected solutions, task IDs, provider generation IDs, keys, rationales, digests, or file paths.

Provisional · Deterministic

IFRS Reporting · 60-task sub-benchmark

Calibration results

Sanitized lifecycle, coverage, score, token, and cost aggregates for FMB-KNOW-ACC-IFRS-02. Results are provisional and specific to the recorded model configuration.

Public Calibration
Sub-benchmark
IFRS Reporting · FMB-KNOW-ACC-IFRS-02
Models
7 calibration runs
Coverage
60 tasks per model
Deterministic
600 deterministic points

Execution detail

Model runs

Provisional deterministic results use deterministic verification only.

Provisional · Configuration-specific calibration · Single runs; no statistical ranking.

xai · high reasoning

grok-4.5 — open run evidence

x-ai/grok-4.5

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

91.4%

548.7 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

93,694

Cost

$0.2066

google-vertex · high reasoning

gemini-3.5-flash — open run evidence

google/gemini-3.5-flash

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

91.9%

551.3 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

129,586

Cost

$0.6085

alibaba · xhigh reasoning

deepseek-v4-flash — open run evidence

deepseek/deepseek-v4-flash

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

90.6%

543.3 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

140,832

Cost

$0.0292

akashml/fp8 · max reasoning

qwen3.6-35b-a3b — open run evidence

qwen/qwen3.6-35b-a3b

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

90.7%

544.0 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

192,470

Cost

$0.1449

together · high reasoning

nemotron-3-ultra-550b-a55b — open run evidence

nvidia/nemotron-3-ultra-550b-a55b

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

89.7%

538.0 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

63,728

Cost

$0.0579

azure · max reasoning

claude-sonnet-5 — open run evidence

anthropic/claude-sonnet-5

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

90.8%

544.7 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

187,444

Cost

$0.837

morph · max reasoning

minimax-m3 — open run evidence

minimax/minimax-m3

StatusProvisional · Deterministic
Progress

60 / 60

Deterministic

88.5%

531.0 / 600

Valid

60 valid answers

0 invalid JSON responses

Provider

0 provider failures

Tokens

69,462

Cost

$0.0497