Financial Analysis · Fifth Order
Guidance Reconciliation
Models reconcile reported results and management guidance across a collection of earnings releases, filings and transcripts. Evaluation checks consistent fiscal periods and units, distinguishes actuals from forward ranges, tracks changed guidance and verifies midpoint arithmetic against cited evidence. A rubric score combines numerical reconciliation and evidence support across 40 cases; higher scores indicate more complete, defensible trackers.
Model results
28 configurations- 189.2%
GPT 6.1 Sol Max reasoning - 287.5%
Muse Wonder 1.0 Max reasoning - 386.1%
Claude Opus 5.5 Adaptive/Max - 485.6%
Grok 4.7 xHigh reasoning - 584.8%Claude Sonnet 5.5 Adaptive/Max
- 684.4%Claude Fable 5.1 Adaptive/Max
- 783.2%GPT 6 Astra Max reasoning
- 883.1%Claude Fable 5 Adaptive/Max
- 982.6%
Gemini 4 Argon High reasoning - 1081.9%Claude Opus 5 Adaptive/Max
- 1181.4%
GLM 5.3 Max reasoning - 1280.3%GPT 6 Sol Max reasoning
- 1379.8%Muse Spark 1.3 Max reasoning
- 1479.6%Muse Spark 1.2 xHigh reasoning
- 1579.2%GPT 5.6 Sol Max reasoning
- 1678.8%Grok 4.6 xHigh reasoning
- 1778.3%
Hy4 Preview High reasoning - 1877.1%
Kimi K3 Max reasoning - 1976.7%
Qwen 3.8 Max xHigh reasoning - 2076.3%Claude Sonnet 5 Adaptive/Max
- 2175.9%
DeepSeek V4.1 Flash 1017 Max reasoning - 2275.6%Gemini 3.7 Flash High reasoning
- 2374.5%DeepSeek V4 Pro Max reasoning
- 2474.2%GLM 5.3 Flash Max reasoning
- 2573.8%Gemini 3.8 Flash High reasoning
- 2672.8%GPT 6 Luna Max reasoning
- 2771.2%
Mistral Large 4 High reasoning - 2870.7%GPT 5.6 Luna Max reasoning
Numeric table · 28 configurations
| Rank | Model / setting | Score |
|---|---|---|
| 1 | GPT 6.1 Sol Max reasoning | 89.2% |
| 2 | Muse Wonder 1.0 Max reasoning | 87.5% |
| 3 | Claude Opus 5.5 Adaptive/Max | 86.1% |
| 4 | Grok 4.7 xHigh reasoning | 85.6% |
| 5 | Claude Sonnet 5.5 Adaptive/Max | 84.8% |
| 6 | Claude Fable 5.1 Adaptive/Max | 84.4% |
| 7 | GPT 6 Astra Max reasoning | 83.2% |
| 8 | Claude Fable 5 Adaptive/Max | 83.1% |
| 9 | Gemini 4 Argon High reasoning | 82.6% |
| 10 | Claude Opus 5 Adaptive/Max | 81.9% |
| 11 | GLM 5.3 Max reasoning | 81.4% |
| 12 | GPT 6 Sol Max reasoning | 80.3% |
| 13 | Muse Spark 1.3 Max reasoning | 79.8% |
| 14 | Muse Spark 1.2 xHigh reasoning | 79.6% |
| 15 | GPT 5.6 Sol Max reasoning | 79.2% |
| 16 | Grok 4.6 xHigh reasoning | 78.8% |
| 17 | Hy4 Preview High reasoning | 78.3% |
| 18 | Kimi K3 Max reasoning | 77.1% |
| 19 | Qwen 3.8 Max xHigh reasoning | 76.7% |
| 20 | Claude Sonnet 5 Adaptive/Max | 76.3% |
| 21 | DeepSeek V4.1 Flash 1017 Max reasoning | 75.9% |
| 22 | Gemini 3.7 Flash High reasoning | 75.6% |
| 23 | DeepSeek V4 Pro Max reasoning | 74.5% |
| 24 | GLM 5.3 Flash Max reasoning | 74.2% |
| 25 | Gemini 3.8 Flash High reasoning | 73.8% |
| 26 | GPT 6 Luna Max reasoning | 72.8% |
| 27 | Mistral Large 4 High reasoning | 71.2% |
| 28 | GPT 5.6 Luna Max reasoning | 70.7% |
Methodology & reference example
Version finance-index-v1 · October 1–28, 2026 · 40 cases per component. Four equally weighted rubric criteria produce a 0–100 score: periods and units; actuals versus outlook; range and midpoint reconciliation; evidence support. The Credo packet below is an authored reference example, not a model run or the source of these cohort scores.
Task instructions
Using only the source packet available at the cutoff, produce a period/metric/reported-actual/guidance table. Reconcile the guided range against the reported quarter, label every derived quantity, cite the sources, and keep the unreported quarter's outcome open. Do not infer realized results or fill missing fields with invented data.
Credo Technology Group packet
Public results release · June 1, 2026 · historical cutoff June 11, 2026 · USD millions unless stated
Credo FY2026 results · June 1, 2026
Company release · June 1, 2026. Q4 FY2026 (ended May 2, 2026) revenue of $437.0 million, up 7.4% sequentially and up 157.0% year over year. Q4 non-GAAP gross margin 68.3% (GAAP 68.2%). Q4 non-GAAP diluted EPS $1.16 (GAAP $0.88). FY2026 revenue $1,335.116 million versus $436.775 million in FY2025. Q1 FY2027 outlook: revenue $465–475 million; non-GAAP gross margin 67–69%; GAAP gross margin 66.9–68.9%; GAAP and non-GAAP operating expenses.
SEC exhibit mirror · June 1, 2026
SEC-filed exhibit mirroring the June 1, 2026 results release. The same reported figures apply: Q4 FY2026 revenue $437.0 million, non-GAAP gross margin 68.3%, non-GAAP diluted EPS $1.16, and Q1 FY2027 guidance covering revenue of $465–475 million, non-GAAP gross margin of 67–69%, GAAP gross margin of 66.9–68.9%, and GAAP and non-GAAP operating expenses.
Authored reference output
Guidance tracker — Credo Q1 FY2027 outlook
As of the June 11, 2026 cutoff, Q1 FY2027 results do not exist, so no guidance outcome is scored. The tracker records what was guided, its basis, and the checkpoint that will resolve it.
| Metric | Q4 FY2026 actual | Q1 FY2027 guidance | Basis / date | Next checkpoint |
|---|---|---|---|---|
| Revenue (USD m) | 437.0 | 465–475 | Guidance · June 1, 2026 | Q1 FY2027 report |
| Revenue midpoint vs Q4 | — | +7.55% (derived) | Analyst arithmetic on guidance | Q1 FY2027 report |
| Non-GAAP gross margin | 68.3% | 67–69% | Guidance · June 1, 2026 | Q1 FY2027 report |
| GAAP gross margin | 68.2% | 66.9–68.9% | Guidance · June 1, 2026 | Q1 FY2027 report |
| Non-GAAP diluted EPS | $1.16 | Not guided | Reported actual only | — |
Guidance is a management outlook, not an actual. Rows without guided values stay empty rather than being extrapolated.
Reading the tracker
Reported actuals, management guidance and derived arithmetic are kept in separate columns. The derived +7.55% midpoint comparison is analyst arithmetic on the guided range, not a company statement.
Credo FY2026 results · June 1, 2026
Resolution
This tracker resolves only when Credo reports Q1 FY2027. Until then, treating the guide as achieved — or as missed — would be an assumption, and this cutoff contains no later information.
Credo FY2026 results · June 1, 2026 · SEC exhibit mirror · June 1, 2026
Criterion-level review approach
- Periods & units
- Keep fiscal periods consistent; distinguish USD millions, percentages and percentage-point changes, and never mix GAAP with non-GAAP bases.
- Reported vs guidance vs derived
- Separate reported actuals from management guidance and from analyst arithmetic such as midpoints; the Q1 FY2027 outlook is guidance, not a later actual.
- Numerical reconciliation
- Reconcile the $465–475m range, the $470m midpoint and the derived +7.55% against the reported $437.0m quarter at full precision before rounding.
- Source support & required fields
- Trace every field to the June 1, 2026 release or its SEC mirror; do not add segment detail or an unsupported full-year forecast.
- Unresolved outcomes
- Leave the guidance outcome open: no actual exists at the June 11, 2026 cutoff, so nothing is scored as achieved or missed.