April 21, 2026

Will a ceasefire or armistice be declared in Ukraine in 2026?

Data for this map is taken from the Ukrainian OSINT DeepState project, a live map of Russian and Ukrainian military operations maintained since April 2022.

Data for this map is taken from the Ukrainian OSINT DeepState project, a live map of Russian and Ukrainian military operations maintained since April 2022.

Phillip Godzin’s pgodzinai forecasts 68%

This is the full report. Condensed version here.

Perplexity response:

# Analysis of Ukraine Ceasefire/Armistice Prospects for 2026

I cannot provide a comprehensive answer based on the search results provided. The results are primarily YouTube video titles without substantive content, and they do not contain detailed information about Russia-Ukraine ceasefire negotiations, peace talks, diplomatic efforts, or expert assessments of ceasefire likelihood in 2026.

To properly address this forecasting question, I would need access to:

Essential sources for context:

  • Recent reporting on Russia-Ukraine peace negotiations and diplomatic channels
  • Current statements from Russian and Ukrainian government officials on ceasefire conditions
  • Analysis of military dynamics and battlefield momentum
  • Expert assessments from conflict analysts and policy institutions
  • International diplomatic efforts (UN, OSCE, neutral countries’ initiatives)
  • Russian economic and military capacity assessments
  • Ukrainian military and political positions on potential peace terms

The search results mention “New progress in US-Iran 45-day ceasefire talks” and various Ukrainian military operations, but these do not directly address Russia-Ukraine ceasefire prospects. To provide you with the detailed rundown of relevant news that would inform your forecast, I would need search results containing substantive reporting on actual Russia-Ukraine peace negotiations, recent diplomatic developments, and expert commentary on ceasefire feasibility for 2026.

Could you provide additional search results or allow me to search for current information on Russia-Ukraine peace negotiations, official statements from both sides regarding ceasefire conditions, and recent diplomatic developments?

Grok response:

  1. Title: Russia’s Putin announces Orthodox Easter ceasefire, expects Ukraine to do same
    Description: Reuters article on Putin’s announcement of a temporary ceasefire for Orthodox Easter.
    Published: April 9, 2026
    Summary: Russian President Vladimir Putin announced a 32-hour ceasefire over Orthodox Easter weekend, from 4 p.m. on April 11 to midnight on April 12, 2026, halting military operations. He expected Ukraine to reciprocate, amid ongoing war. This follows Zelenskyy’s earlier calls for Easter truces, though no broader peace deal has emerged.[1][2]
  2. Title: Russia and Ukraine agree to truce for Orthodox Easter
    Description: BBC report confirming mutual agreement on the Easter ceasefire.
    Published: April 10, 2026
    Summary: Both Russia and Ukraine agreed to a temporary truce for Orthodox Easter, starting April 11, as proposed by Zelenskyy and accepted after Putin’s order. Kyiv seeks a full ceasefire as a step to negotiations, but Russia insists on permanent peace terms. Skepticism persists due to past violations of short truces.[2]
  3. Title: Putin declares a ceasefire in Ukraine for Orthodox Easter
    Description: AP News coverage of the declaration.
    Published: April 10, 2026
    Summary: Putin ordered a 32-hour halt to fighting over Easter, following Zelenskyy’s proposal, marking a rare mutual pause. Ukrainian officials welcomed it but emphasized vigilance. No extension or broader ceasefire was indicated.[3]
  4. Title: Zelenskiy, welcoming Iran de-escalation, says Ukraine ready for ceasefire with Russia
    Description: Reuters on Zelenskyy’s ceasefire readiness amid other global events.
    Published: April 8, 2026
    Summary: Zelenskiy reiterated Ukraine’s call for a full ceasefire to enable peace talks, amid de-escalation elsewhere like Iran. Russia has not committed beyond short truces. This reflects ongoing diplomatic pushes without resolution.[4]
  5. Title: Moscow says Ukraine peace talks on pause, Kyiv says new proposals incoming
    Description: Reuters on stalled negotiations.
    Published: March 19, 2026
    Summary: Kremlin stated US-Russia-Ukraine talks are paused due to external factors like Iran tensions. Kyiv insists on new proposals and rejects territorial concessions. No ceasefire emerged from these efforts.[5]
  6. Title: 2026 United States–Ukraine–Russia meetings in Geneva
    Description: Wikipedia entry on early 2026 trilateral talks.
    Published: Ongoing, events Feb 17-18, 2026
    Summary: US-brokered talks in Geneva in February 2026 failed to produce a ceasefire, with disagreements on security and territory. Russia pushed maximalist demands; Ukraine sought guarantees. Talks stalled shortly after.[6]
  7. Title: Russia x Ukraine ceasefire by end of 2026?
    Description: Polymarket prediction market odds.
    Updated: April 10, 2026
    Summary: Market prices “Yes” at 30% for any ceasefire before 2027, reflecting low crowd expectation for sustained halt. Earlier markets for Q1 2026 resolved No (0%). Easter truce not seen as qualifying for these markets.[7]
  8. Title: Russian Offensive Campaign Assessment, April 8, 2026
    Description: ISW update on frontline.
    Published: April 8, 2026
    Summary: Russian advances continue near several fronts; no broad pause observed pre-Easter announcement. Ukraine leverages drones effectively. Fighting persists into April despite diplomatic rhetoric.[8]
  9. Title: Ukraine agrees to multi-tiered ceasefire enforcement plan with Europe and US
    Description: Reuters on potential future mechanisms.
    Published: February 2, 2026
    Summary: Ukraine agreed to enforcement plans for any future ceasefire, involving monitors. No actual ceasefire implemented then. Russia rejected interim halts without full concessions.[9]
  10. Title: Russia–Ukraine War in 2026: At the Crossroads of Peace and War
    Description: ORF analysis on 2026 outlook.
    Published: February 26, 2026
    Summary: Predicts possible fragile ceasefire under Trump diplomacy, but Russia’s demands clash with Ukraine’s needs. Economy and mercenaries strain Russia, per background. No armistice likely without major shifts.[10]

Current Status (as of April 10, 2026)

No armistice or sustained ceasefire covering the entire front has been declared in 2026, despite multiple diplomatic efforts. The most recent development is a mutual 32-hour temporary truce for Orthodox Easter (April 11-12), announced by Putin on April 9 and welcomed by Zelenskyy, reported widely by major outlets like Reuters, BBC, and AP.[1][2][3] This aligns with the background’s definition of a temporary ceasefire but is limited in scope and duration, with both sides maintaining defensive postures. Prior short-term arrangements, like energy infrastructure halts in February (Zelenskiy noted Russian observance then) and local Zaporizhzhia truces, have not escalated to broader pauses.[11][12]

Diplomatic Efforts and Talks

Early 2026 saw US-brokered trilateral meetings in Geneva (Feb 17-18), focusing on de-escalation but yielding no agreement due to Russia’s demands for Ukrainian neutrality and territorial recognition, which Kyiv rejected.[6][13] Talks paused by mid-March amid Russia’s spring offensive and distractions like Iran tensions, with Moscow calling them a “situational pause” and Kyiv preparing new proposals.[5][14] Zelenskiy has repeatedly proposed full ceasefires (e.g., April 6, contingent on Russian reciprocity) tied to peace negotiations, but Putin insists on comprehensive deals, not interim halts.[15] Allies urged unconditional ceasefires on the war’s fourth anniversary (Feb 24), but no progress.[16]

Military Situation

As of April 8-9, Russian forces advanced near Siversk and other fronts, with drone strikes on civilians and infrastructure continuing pre-Easter announcement.[8][17] Ukraine reports desertions from fatigue but maintains drone superiority.[18] No signs of culmination; Russia recruits mercenaries globally, economy strained by sanctions (higher oil prices help amid Iran issues).[19]

Expert and Market Views

Polymarket odds for sustained ceasefires remain low: 30% by end-2026, 2-10% for Q2 deadlines, post-Easter truce not altering short-term markets.[7][20] Analysts note Trump’s June deadline rhetoric but stalled progress; Russia weighs halting talks without concessions.[21][22] ORF (Feb) highlights fragile ceasefire potential but maximalist Russian stance.[10] Recent X posts reflect skepticism on enforcement.[23]

No authoritative resolution source (e.g., UN) confirms broader halt; major outlets report ongoing war with temporary pauses only.[24]

Extracted resolution criteria links:
[]

Extracted resolution criteria links:
[]


Anthropic Binary response:

Reasoning

Key Development: Easter Ceasefire Already Declared

The most critical piece of evidence is that on April 9, 2026, Putin officially declared a 32-hour ceasefire for Orthodox Easter (April 11-12), and Ukraine agreed on April 10. This is reported by Reuters, BBC, and AP — all major news outlets — using the explicit word “ceasefire.”

Resolution Criteria Analysis:
The criteria requires: “a ceasefire or armistice is officially declared between Russia and Ukraine at any point in 2026, as reported by major news outlets.”

Checking each element:

  • ✅ Officially declared (Putin’s order, Zelenskyy’s acceptance)
  • ✅ Between Russia and Ukraine
  • ✅ In 2026 (ceasefire takes effect April 11, after today)
  • ✅ Reported by major news outlets (Reuters, BBC, AP all use “ceasefire”)
  • ✅ Background explicitly states: “A ceasefire is temporary or might only cover one region”

Critical Assumptions to Stress-Test:

  1. Will the question operator interpret “ceasefire” literally? (~65% yes) The literal text and background note support counting temporary ceasefires. However, the spirit of the question may aim at something more substantial. Polymarket prices a “sustained ceasefire” at only 30%, suggesting markets distinguish between holiday truces and meaningful ceasefires.
  2. Does the timing requirement exclude the Easter ceasefire? (~15% risk) The declaration was April 9-10, but the ceasefire itself takes effect April 11 (after today). The natural reading is that the ceasefire happening after today satisfies this.
  3. If Easter ceasefire doesn’t count, will a more significant ceasefire occur? (~30-35% probability) Given active US-brokered diplomacy, Zelenskyy’s repeated ceasefire proposals, economic strain on Russia, and 8+ months remaining, there’s meaningful probability of a broader ceasefire.

Combined Calculation:

  • P(Easter ceasefire qualifies) ≈ 60%
  • P(another ceasefire if Easter doesn’t count) ≈ 32%
  • Combined: 0.60 + 0.40 × 0.32 ≈ 0.73

Adjusting for historical overconfidence bias and the signal from prediction markets (which seem to not count short truces): slight downward adjustment.

Probability: 72%
The extracted probability is: 72%


Anthropic Binary response:

Reasoning

Key Evidence Review

The Easter Ceasefire (April 11-12, 2026)
The most critical development: Putin announced a 32-hour ceasefire on April 9, Ukraine agreed on April 10, and the ceasefire takes effect April 11-12. This is reported by Reuters, BBC, and AP — all major outlets.

Resolution Criteria Check:

  • “Officially declared” ✓ (both sides announced it)
  • “Between Russia and Ukraine” ✓
  • “At any point in 2026” ✓
  • “Reported by major news outlets” ✓
  • The background explicitly states: “A ceasefire is temporary or might only cover one region of the front” — suggesting temporary ceasefires qualify

Timing Constraint: “The event must happen after today (April 10).” The ceasefire takes effect April 11, which is after today. The declaration itself was April 9-10, creating some ambiguity.

Critical Assumptions to Stress-Test

Does the Easter ceasefire qualify as “a ceasefire” under resolution criteria? (Most impactful)

  • The background definition explicitly includes temporary ceasefires
  • However, Polymarket at 30% suggests their markets don’t count short truces (though their criteria may differ)
  • Probability this counts: ~65%

Does the timing work? The ceasefire was declared April 9-10 but implemented April 11-12.

  • The ceasefire period begins after today → likely qualifies
  • Risk of exclusion: ~15%

Could the ceasefire collapse before implementation?

  • Past truces have been violated, but an official declaration still occurred
  • Risk: ~5%

Probability Calculation

  • P(Easter ceasefire qualifies and resolves Yes) ≈ 65% × 85% × 95% ≈ 52%
  • P(some other ceasefire in remaining ~8.5 months if Easter doesn’t count) ≈ 35%
  • Geneva talks could resume; ongoing diplomatic pressure from US; Trump’s stated deadlines
  • Multiple holidays/occasions for short truces
  • Combined P(Yes) ≈ 0.52 + (0.48 × 0.35) ≈ 0.69

Adjustments

  • Historical overconfidence bias: slight downward adjustment
  • The fact that previous truces (energy infrastructure halt, Zaporizhzhia) apparently didn’t resolve this question suggests some possibility the Easter ceasefire also won’t count
  • But this question has a “must happen after today” constraint, meaning previous truces were excluded by design, not by type

Probability: 68%
The extracted probability is: 68%

Gemini Binary response:

This forecast is predicated on a careful analysis of the resolution criteria, current diplomatic and military realities, and recent precedents.

Analysis of the Forecasting Question

The question is whether a ceasefire or armistice will be officially declared in Ukraine between today, April 10, 2026, and the end of the year. The resolution criteria are broad and do not specify a minimum duration or geographical scope for the ceasefire. This means that any officially declared pause in fighting, even if temporary or localized, would resolve the question as “Yes”.

Key Factors Influencing the Forecast

  1. Recent Precedent of Temporary Ceasefires: The most significant factor is the recently declared 32-hour ceasefire for Orthodox Easter (April 11-12, 2026). This event, occurring at the very start of the forecast window, demonstrates that despite deep-seated disagreements, both Russia and Ukraine are currently capable and willing to declare and observe short-term, temporary truces for specific occasions. This sets a strong precedent for similar events happening again in 2026.
  2. Distinction Between Ceasefire Types: It is crucial to differentiate between two potential paths to a “Yes” resolution:
    • Major Armistice or Sustained Ceasefire: The probability of a comprehensive, lasting cessation of hostilities remains low. High-level talks have consistently stalled due to Russia’s maximalist demands—including territorial concessions and Ukrainian neutrality—which Kyiv categorically rejects. Expert analysis and prediction markets, which are likely pricing in a more substantial halt to the war, reflect this pessimism, with odds for a deal by year-end remaining low (e.g., 30% on Polymarket). The chance of this specific outcome in the next eight months is likely below 10%.
    • Temporary or Localized Ceasefire: The probability of another short-term truce is considerably higher. The Easter ceasefire provides a template. Another religiously significant holiday, such as Christmas, presents a clear opportunity for a repeat. Furthermore, the war has previously seen localized truces declared for specific purposes, such as enabling repairs at the Zaporizhzhia nuclear plant. Over the remaining eight months of the year, it is plausible that another acute humanitarian crisis, the need for a large-scale prisoner exchange, or another critical infrastructure issue could prompt a similar declared pause.

Forecasting Rationale

The forecast hinges on the high likelihood of a temporary ceasefire being declared, rather than a full-fledged armistice. While a breakthrough in peace negotiations appears unlikely in 2026, the precedent for temporary pauses is now firmly established.

  • Holiday Ceasefire: A declared ceasefire around the Christmas holiday period is a plausible scenario, mirroring the Easter truce.
  • Humanitarian/Special Purpose Ceasefire: The dynamics of the ongoing conflict make it probable that a situation will arise requiring a declared, localized truce to allow for civilian evacuation, disaster response, or prisoner exchanges.

My final probability is an aggregation of the chances of these different types of events occurring. The primary driver for a “Yes” resolution is the strong possibility of another temporary truce, for which there is now a direct and immediate precedent. The low probability of a major armistice is a minor component of the overall forecast but still contributes to the “Yes” side.

Key Assumptions

  1. Broad Interpretation of “Ceasefire”: My forecast assumes that any officially declared pause in fighting, regardless of its duration or scope (e.g., a 24-hour Christmas truce), will satisfy the resolution criteria. The failure of this assumption (i.e., if an unstated minimum duration is required) would drastically lower the forecast probability. I estimate the probability of this assumption being incorrect as low (<5%).
  2. Easter Truce as a Valid Precedent: I assume the Easter 2026 truce is not a unique anomaly and that the conditions allowing for it could be replicated later in the year. A significant battlefield shift or a complete breakdown in even informal communication channels could invalidate this assumption. I estimate the probability of this assumption being incorrect as moderate (around 40%).

Taking these factors and the high potential for another temporary truce into account, the probability of a “Yes” resolution is above 50%. While a lasting peace is not in sight, a declared pause in fighting is a much lower and more achievable bar.

Probability: 51%

The extracted probability is: 51%


Anthropic Binary response:

Analyzing Ukraine Ceasefire/Armistice Probability for 2026

Key Development: Easter Ceasefire Already Declared

The most critical piece of evidence is that on April 9-10, 2026, Putin declared a 32-hour ceasefire for Orthodox Easter (April 11-12), and Ukraine officially agreed. This was widely reported by Reuters, BBC, and AP News.

Does the Easter Ceasefire Already Resolve the Question?

Resolution criteria: “Resolves Yes if either a ceasefire or armistice is officially declared between Russia and Ukraine at any point in 2026, as reported by major news outlets.”

Background notes: “A ceasefire is temporary or might only cover one region of the front.”

The Easter ceasefire is:

  • ✅ Officially declared by both sides
  • ✅ Between Russia and Ukraine
  • ✅ In 2026
  • ✅ Reported by major news outlets (Reuters, BBC, AP)

However, the question states “The event must happen after today [April 10].” The declaration was made April 9-10, but the ceasefire itself takes effect April 11 at 4 PM. This creates interpretive ambiguity.

Key Assumptions and Failure Probabilities

Assumption 1: The 32-hour Easter truce qualifies as “a ceasefire officially declared”

  • Probability correct: ~75% (the literal text and background support this; counterargument is the question may intend something more substantial)
  • If wrong: need a broader ceasefire

Assumption 2: The ceasefire taking effect April 11 satisfies “after today”

  • Probability correct: ~60% (declaration was April 9-10, but implementation is April 11; strict interpretation could exclude it)

Assumption 3: If Easter ceasefire doesn’t count, a subsequent ceasefire has ~40% chance by year-end

  • Based on: Polymarket at 30% for sustained ceasefire, plus additional probability of more temporary truces (Easter precedent makes future holiday truces more likely), 8+ months remaining, ongoing US diplomatic pressure

Probability Calculation

  • P(Easter ceasefire satisfies all criteria) = 0.75 × 0.60 = 0.45
  • P(ceasefire doesn’t collapse before taking effect) = 0.95
  • P(Easter resolves Yes) = 0.45 × 0.95 = 0.43
  • P(new ceasefire declared after April 10 | Easter doesn’t count) = 0.40
  • P(Yes from future ceasefire) = 0.57 × 0.40 = 0.23
  • P(Yes total) = 0.43 + 0.23 = 0.66

Calibration Check

  • Polymarket: 30% for “sustained” ceasefire (stricter criteria than this question)
  • This question’s criteria is more lenient (temporary ceasefires explicitly count)
  • Historical overconfidence adjustment: slight downward correction

The Easter ceasefire creates substantial probability mass, and even if it doesn’t count, the precedent of temporary truces plus ongoing diplomatic efforts over 8+ months provide meaningful additional probability.

Probability: 66%
The extracted probability is: 66%

GPT Binary response:

The key issue is that this question is easier than “will the war end?” but harder than it first looks.

A few things matter most:

  • Resolution standard is relatively loose. It does not require a peace treaty, durable compliance, or a front-wide permanent halt. A temporary ceasefire, even regional, can count if it is officially declared between Russia and Ukraine and reported by major outlets.
  • But the event must happen after today. The just-announced Orthodox Easter truce strongly suggests such declarations are possible, but I am treating that specific truce as not counting for this forecast because it appears to have already been declared on or before April 10.
  • The default path is still continued war. Reuters/BBC/AP reporting and ISW-type battlefield analysis all point to ongoing fighting, stalled talks, and large gaps between Russian and Ukrainian positions. Russia still appears to think it can improve its position militarily; Ukraine still rejects the terms Russia wants. That makes a real armistice or broad ceasefire hard.

How I weight the evidence:

  1. Most important: Reuters/BBC/AP recent reporting.
    These are the highest-value inputs. They show:
  • a real, mutual Easter truce was possible;
  • Kyiv remains open to a fuller ceasefire;
  • Moscow still rejects an interim halt absent major concessions;
  • talks have stalled rather than accelerated. Net effect: raises odds above very low levels, but not to even money.
  1. Second most important: battlefield and bargaining incentives.
    As of early April, Russia is still pressing offensives. States that think time is on their side usually do not agree to meaningful armistices soon. This is the strongest argument for a “No.”
  2. Third: market/implied crowd priors.
    The cited prediction market around 30 for ceasefire before 2027 is useful, though imperfect because its rules may be stricter than this market’s. Since this question seems a bit easier to satisfy than a “substantive peace deal,” I should be somewhat above that, but not massively.

My rough decomposition from now to year-end:

  • Another short, symbolic, or localized ceasefire/truce: modestly plausible. The Easter example shows the channel exists. There are multiple future windows: humanitarian pauses, infrastructure repair, nuclear safety, prisoner exchange-linked pauses, holiday pauses, or a limited front-sector halt.
  • A broader ceasefire as part of renewed talks: possible, but still clearly the less likely branch. It would likely require either Russian offensive culmination, stronger outside pressure, or a sharper deterioration in Russia’s economic/military position.
  • Full armistice across the whole war: low probability within this timeframe.

If I sketch the timelines:

  • By end of June: low chance beyond the current Easter episode.
  • By end of September: somewhat higher if Russia’s current push stalls.
  • By end of December: highest, because exhaustion, winter infrastructure risks, and diplomatic pressure accumulate.

That pattern supports a number in the low-to-mid 30s rather than the teens or 50+.

The assumptions most likely to be wrong and most important:

  1. Assumption: Russia remains unwilling to accept even a limited new ceasefire while it believes it has battlefield leverage.
    Chance this assumption is wrong: about 30.
    If wrong, odds rise materially.
  2. Assumption: A future short/local truce comparable to the Easter pause does not become common enough to make “Yes” likely.
    Chance this assumption is wrong: about 25.
    If wrong, this market could resolve “Yes” on a relatively small event.
  3. Assumption: The already announced Easter ceasefire will not be used to resolve this market.
    Chance this assumption is wrong: small, around 5.
    If wrong, the true forecasting problem is mostly already over.

Balancing all of that, I think the base case is still no new officially declared ceasefire/armistice after today, but the bar for a “Yes” is low enough, and the Easter precedent recent enough, that the probability is meaningfully above a generic peace-deal forecast.

Probability: 36%

The extracted probability is: 36%
Task failed with error:
Traceback (most recent call last):
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 84, in fetch_gpt_prediction
return await fetch_sync(get_gpt_binary_prediction, title, perplexity_summary_report, asknews_summary, asknews_deep_research, vision_analysis, grok_report, adjacent_markets_research, today, resolution_date, background, fine_print, resolution_criteria)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 74, in fetch_sync
return await asyncio.get_event_loop().run_in_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TimeoutError

Task failed with error: ‘choices’
Traceback (most recent call last):
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 84, in fetch_gpt_prediction
return await fetch_sync(get_gpt_binary_prediction, title, perplexity_summary_report, asknews_summary, asknews_deep_research, vision_analysis, grok_report, adjacent_markets_research, today, resolution_date, background, fine_print, resolution_criteria)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 74, in fetch_sync
return await asyncio.get_event_loop().run_in_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/thread.py”, line 58, in run
result = self.fn(*self.args, *self.kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 70, in run_sync return future.result(timeout=180) ^^^^^^^^^^^^^^^^^^^^^^^^^^ File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/_base.py”, line 456, in result return self.__get_result() ^^^^^^^^^^^^^^^^^^^ File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/_base.py”, line 401, in __get_result raise self._exception File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/thread.py”, line 58, in run result = self.fn(self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/openai_service.py”, line 121, in get_gpt_binary_prediction
text = chat_completion[“choices”][0][“message”][“content”]
~~~^^^^^^^^^^^
KeyError: ‘choices’

Task failed with error: ‘choices’
Traceback (most recent call last):
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 84, in fetch_gpt_prediction
return await fetch_sync(get_gpt_binary_prediction, title, perplexity_summary_report, asknews_summary, asknews_deep_research, vision_analysis, grok_report, adjacent_markets_research, today, resolution_date, background, fine_print, resolution_criteria)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 74, in fetch_sync
return await asyncio.get_event_loop().run_in_executor(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/thread.py”, line 58, in run
result = self.fn(*self.args, *self.kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 70, in run_sync return future.result(timeout=180) ^^^^^^^^^^^^^^^^^^^^^^^^^^ File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/_base.py”, line 456, in result return self.__get_result() ^^^^^^^^^^^^^^^^^^^ File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/_base.py”, line 401, in __get_result raise self._exception File “/opt/hostedtoolcache/Python/3.11.15/x64/lib/python3.11/concurrent/futures/thread.py”, line 58, in run result = self.fn(self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/openai_service.py”, line 121, in get_gpt_binary_prediction
text = chat_completion[“choices”][0][“message”][“content”]
~~~^^^^^^^^^^^
KeyError: ‘choices’

All predictions: [51, 66, 68, 72]
Filtered predictions: [66, 68]

Median reasoning:

Reasoning

Key Evidence Review

The Easter Ceasefire (April 11-12, 2026)
The most critical development: Putin announced a 32-hour ceasefire on April 9, Ukraine agreed on April 10, and the ceasefire takes effect April 11-12. This is reported by Reuters, BBC, and AP — all major outlets.

Resolution Criteria Check:

  • “Officially declared” ✓ (both sides announced it)
  • “Between Russia and Ukraine” ✓
  • “At any point in 2026” ✓
  • “Reported by major news outlets” ✓
  • The background explicitly states: “A ceasefire is temporary or might only cover one region of the front” — suggesting temporary ceasefires qualify

Timing Constraint: “The event must happen after today (April 10).” The ceasefire takes effect April 11, which is after today. The declaration itself was April 9-10, creating some ambiguity.

Critical Assumptions to Stress-Test

Does the Easter ceasefire qualify as “a ceasefire” under resolution criteria? (Most impactful)

  • The background definition explicitly includes temporary ceasefires
  • However, Polymarket at 30% suggests their markets don’t count short truces (though their criteria may differ)
  • Probability this counts: ~65%

Does the timing work? The ceasefire was declared April 9-10 but implemented April 11-12.

  • The ceasefire period begins after today → likely qualifies
  • Risk of exclusion: ~15%

Could the ceasefire collapse before implementation?

  • Past truces have been violated, but an official declaration still occurred
  • Risk: ~5%

Probability Calculation

  • P(Easter ceasefire qualifies and resolves Yes) ≈ 65% × 85% × 95% ≈ 52%
  • P(some other ceasefire in remaining ~8.5 months if Easter doesn’t count) ≈ 35%
  • Geneva talks could resume; ongoing diplomatic pressure from US; Trump’s stated deadlines
  • Multiple holidays/occasions for short truces
  • Combined P(Yes) ≈ 0.52 + (0.48 × 0.35) ≈ 0.69

Adjustments

  • Historical overconfidence bias: slight downward adjustment
  • The fact that previous truces (energy infrastructure halt, Zaporizhzhia) apparently didn’t resolve this question suggests some possibility the Easter ceasefire also won’t count
  • But this question has a “must happen after today” constraint, meaning previous truces were excluded by design, not by type

Probability: 68%

=== Forecast Result ===


{
“probability_yes”: 0.67,
“probability_yes_per_category”: null,
“continuous_cdf”: null
}

=== Reasoning ===

Reasoning

Key Evidence Review

The Easter Ceasefire (April 11-12, 2026)
The most critical development: Putin announced a 32-hour ceasefire on April 9, Ukraine agreed on April 10, and the ceasefire takes effect April 11-12. This is reported by Reuters, BBC, and AP — all major outlets.

Resolution Criteria Check:

  • “Officially declared” ✓ (both sides announced it)
  • “Between Russia and Ukraine” ✓
  • “At any point in 2026” ✓
  • “Reported by major news outlets” ✓
  • The background explicitly states: “A ceasefire is temporary or might only cover one region of the front” — suggesting temporary ceasefires qualify

Timing Constraint: “The event must happen after today (April 10).” The ceasefire takes effect April 11, which is after today. The declaration itself was April 9-10, creating some ambiguity.

Critical Assumptions to Stress-Test

Does the Easter ceasefire qualify as “a ceasefire” under resolution criteria? (Most impactful)

  • The background definition explicitly includes temporary ceasefires
  • However, Polymarket at 30% suggests their markets don’t count short truces (though their criteria may differ)
  • Probability this counts: ~65%

Does the timing work? The ceasefire was declared April 9-10 but implemented April 11-12.

  • The ceasefire period begins after today → likely qualifies
  • Risk of exclusion: ~15%

Could the ceasefire collapse before implementation?

  • Past truces have been violated, but an official declaration still occurred
  • Risk: ~5%

Probability Calculation

  • P(Easter ceasefire qualifies and resolves Yes) ≈ 65% × 85% × 95% ≈ 52%
  • P(some other ceasefire in remaining ~8.5 months if Easter doesn’t count) ≈ 35%
  • Geneva talks could resume; ongoing diplomatic pressure from US; Trump’s stated deadlines
  • Multiple holidays/occasions for short truces
  • Combined P(Yes) ≈ 0.52 + (0.48 × 0.35) ≈ 0.69

Adjustments

  • Historical overconfidence bias: slight downward adjustment
  • The fact that previous truces (energy infrastructure halt, Zaporizhzhia) apparently didn’t resolve this question suggests some possibility the Easter ceasefire also won’t count
  • But this question has a “must happen after today” constraint, meaning previous truces were excluded by design, not by type

Probability: 68%

End